Clean public docs language and promote diagram render smoke CI to required

Author: CameronBrooks11

.github/workflows/ci.yml

@@ -198,9 +198,8 @@ jobs:
         run: ctest --preset ci-windows-release
 
   diagram-render-smoke:
-    name: Diagram Render Smoke (advisory)
+    name: Diagram Render Smoke
     runs-on: ubuntu-24.04
-    continue-on-error: true
 
     steps:
       - name: Checkout

CHANGELOG.md

@@ -36,16 +36,16 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - server protected-write rejection
   - server `--dt` timestep behavior
   - SignalStore unit-declaration behavior
-- Benchmark evidence tooling for Phase 2:
+- Benchmark evidence tooling:
   - policy profiles (`local`, `ci-hosted`, `ci-dedicated`)
   - benchmark evaluation script (`scripts/evaluate_benchmarks.py`)
   - benchmark evidence CI workflow with artifact upload (`.github/workflows/benchmark-evidence.yml`)
-- Phase 3 numerical validation framework:
+- Numerical validation framework:
   - `thermal_mass` integration method policy (`forward_euler`, `rk4`)
   - numerical validation runner (`scripts/run_numerical_validation.py`)
   - thermal validation target (`tests/validation/thermal_mass_validation.cpp`)
   - numerical validation CI evidence workflow (`.github/workflows/numerical-validation-evidence.yml`)
-  - methodology/evidence docs (`docs/validation-methodology.md`, `docs/phase3-evidence.md`)
+  - methodology/evidence docs (`docs/validation-methodology.md`, `docs/numerical-methods.md`)
 
 ### Fixed
 

docs/graph-visualization.md

@@ -1,17 +1,17 @@
-# Graph Visualization Contract
+# Graph Visualization
 
-**Status:** Phase 4 contract; Sprint 4.3 renderer integration implemented.  
-**Purpose:** lock the visualization interface and behavior before coding to prevent drift in tests and CLI behavior.
+**Status:** Implemented  
+**Purpose:** define the public behavior and evidence contract for FluxGraph graph-diagram generation.
 
 ## 1. Scope
 
-This phase adds an optional graph-diagram toolchain that:
+FluxGraph provides an optional graph-diagram toolchain that:
 
-1. takes FluxGraph graph definitions (JSON/YAML via existing loaders, or direct `GraphSpec` in C++),
-2. produces deterministic DOT output as the canonical artifact,
+1. accepts FluxGraph graph definitions (JSON/YAML via existing loaders, or direct `GraphSpec` in C++),
+2. emits deterministic DOT as the canonical artifact,
 3. optionally renders DOT to image formats using external Graphviz CLI tooling.
 
-## 2. Non-Goals (Phase 4)
+## 2. Current Non-Goals
 
 1. Live runtime graph tracing or time-series overlays.
 2. Interactive web UI or editor.
@@ -21,22 +21,18 @@ This phase adds an optional graph-diagram toolchain that:
 ## 3. Build and Dependency Contract
 
 1. Feature flag: `FLUXGRAPH_BUILD_DIAGRAM_TOOL` (default `OFF`).
-2. No new vcpkg feature is required for the CLI-renderer path in Phase 4.
+2. No new vcpkg feature is required for the current CLI-renderer path.
 3. Rendering uses external `dot` process when image output is requested.
 4. Core `fluxgraph` library dependency contract remains unchanged.
 
 ## 4. Architecture Contract
 
-Planned split:
-
 1. `viz-core`: pure C++ emitter (`GraphSpec -> DOT`), no third-party dependencies.
 2. `fluxgraph-diagram` CLI: input parsing/validation, file I/O, optional renderer invocation.
 
 `viz-core` depends on FluxGraph public graph spec types (`GraphSpec`, `ModelSpec`, `EdgeSpec`, `RuleSpec`) and does not depend on runtime engine internals.
 
-## 5. C++ API Sketch (for Direct Integrator Use)
-
-Planned public surface (names may be finalized during implementation, contract intent is stable):
+## 5. C++ API Sketch (Direct Integrator Use)
 
 ```cpp
 #include <fluxgraph/graph/spec.hpp>
@@ -54,12 +50,12 @@ std::string emit_dot(const GraphSpec& spec, const DotEmitOptions& options = {});
 } // namespace fluxgraph::viz
 ```
 
-Contract guarantees:
+Guarantees:
 
 1. `emit_dot` is deterministic for the same `GraphSpec` and options.
 2. Resulting DOT is syntactically valid for Graphviz `dot`.
 
-## 6. DOT Emission Policy (Determinism + Escaping)
+## 6. DOT Emission Policy
 
 ## 6.1 Identifier and Label Escaping
 
@@ -68,7 +64,7 @@ Contract guarantees:
 - `\` -> `\\`
 - `"` -> `\"`
 - newline -> `\n`
-3. Signal paths containing `/`, `.`, spaces, or other punctuation are valid and must be preserved via quoting.
+3. Signal paths containing `/`, `.`, spaces, or other punctuation are preserved via quoting.
 
 ## 6.2 Ordering Rules
 
@@ -82,11 +78,11 @@ No reliance on container insertion order is allowed for canonical DOT output.
 
 1. Unknown transform types are rendered (not dropped) using raw `type` text in edge annotation.
 2. Unknown transform types do not fail DOT generation unless required fields are structurally invalid.
-3. CLI should emit a non-fatal warning for unknown transform types.
+3. CLI emits a non-fatal warning for unknown transform types.
 
-## 7. CLI Contract Phasing
+## 7. CLI Contract
 
-Supported CLI flags:
+Supported flags:
 
 1. `--in <path>`: required graph input (`.json`, `.yaml`, `.yml`)
 2. `--out <path>`: required output artifact path
@@ -101,28 +97,22 @@ Behavior:
 3. Graphviz invocation failures are explicit and include command context.
 4. For image outputs, `--dot-out` (if provided) must differ from `--out`.
 
-Exit code contract (planned):
+Exit codes:
 
 1. `0`: success
 2. `1`: input/load/validation failure
-3. `2`: unsupported option/format for current build
+3. `2`: unsupported option/format
 4. `3`: renderer invocation failure
 
-Current implementation status:
-
-1. DOT emission is implemented.
-2. SVG/PNG rendering via Graphviz CLI is implemented.
-3. `--dot-bin` and `--dot-out` are implemented.
-
-## 8. Evidence and Governance Mapping
-
-Planning source for claim-evidence governance: `working/archive/claim_evidence_matrix.md`.
-
-Visualization claims and expected evidence for this phase:
-
-1. "Deterministic DOT output": `tests/unit/viz_dot_emitter_test.cpp` and `tests/unit/viz_dot_golden_test.cpp`.
-2. "CLI DOT determinism end-to-end": `tests/integration/diagram_cli_e2e.cmake` (registered as `integration.diagram_cli_json_dot`).
-3. "Core isolation": default presets keep `FLUXGRAPH_BUILD_DIAGRAM_TOOL=OFF`.
-4. "Optional rendering": advisory GitHub Actions job `diagram-render-smoke` in `.github/workflows/ci.yml`.
-
-This mapping is the baseline evidence package for the Phase 4 visualization surface.
+## 8. Evidence Mapping
+
+1. Deterministic DOT output:
+- `tests/unit/viz_dot_emitter_test.cpp`
+- `tests/unit/viz_dot_golden_test.cpp`
+2. CLI DOT determinism end-to-end:
+- `tests/integration/diagram_cli_e2e.cmake` (registered as `integration.diagram_cli_json_dot`)
+3. Core isolation:
+- default presets keep `FLUXGRAPH_BUILD_DIAGRAM_TOOL=OFF`
+4. Renderer coverage in CI:
+- `linux-matrix` lane `diagram-dot`
+- `diagram-render-smoke` job in `.github/workflows/ci.yml`

docs/numerical-methods.md

@@ -7,7 +7,7 @@ It complements `docs/semantics_spec.md` and is authoritative for solver
 selection and stability interpretation.
 Validation protocol details are documented in `docs/validation-methodology.md`.
 
-## Current Policy (Phase 3.1 baseline)
+## Current Policy
 
 1. Integration method selection is explicit per model via model parameters.
 2. If not specified, the deterministic default is `forward_euler`.
@@ -43,7 +43,7 @@ If `h <= 0`, the model is treated as unconditionally stable for this criterion.
 
 ## Validation Expectations
 
-Phase 3 validation must report:
+Validation runs must report:
 
 1. Error metrics (`L2`, `Linf`) versus analytical references.
 2. Convergence behavior as `dt` is refined.
@@ -59,7 +59,7 @@ Future methods (for example trapezoidal or implicit schemes) must define:
 
 ## Reproducible Validation Run
 
-Use the validation runner to produce Phase 3 artifacts:
+Use the validation runner to produce validation artifacts:
 
 ```bash
 python scripts/run_numerical_validation.py --preset dev-release --enforce-order

docs/performance-benchmarks.md

@@ -6,7 +6,7 @@ This document defines the reproducible benchmark workflow for FluxGraph.
 
 Goals:
 
-1. Provide repeatable measurement artifacts for Phase 2 gates.
+1. Provide repeatable measurement artifacts for benchmark quality gates.
 2. Separate measured evidence from narrative claims.
 3. Keep benchmark runs deterministic enough for regression comparison.
 
@@ -87,7 +87,7 @@ Required files:
 4. Git commit hash and dirty-worktree flag
 5. Per-benchmark executable path, command, exit code, duration, parsed PASS/FAIL status lines
 
-Tick benchmark output additionally tracks measured heap allocations during the timed loop (`Allocations`, `Alloc/tick`) so Phase 2 zero-allocation evidence is captured per scenario.
+Tick benchmark output additionally tracks measured heap allocations during the timed loop (`Allocations`, `Alloc/tick`) so zero-allocation evidence is captured per scenario.
 
 `benchmark_evaluation.json` contains:
 
@@ -134,7 +134,7 @@ Recommended:
 2. Store artifacts as CI build artifacts.
 3. Apply `ci-hosted` profile on hosted runners and reserve strict gating for dedicated runners.
 
-## Next Phase 2 Steps
+## Next Steps
 
 1. Calibrate thresholds using several hosted-runner samples (reduce false positives while preserving sensitivity).
 2. Provision and commit a true `ci-dedicated` baseline from stable hardware.

docs/semantics_spec.md

@@ -1,7 +1,7 @@
 # FluxGraph Semantics Specification
 
-**Status:** Phase 0 baseline (pending dual sign-off)  
-**Version:** 0.1.0-phase0  
+**Status:** Baseline semantics specification  
+**Version:** 0.1.0  
 **Last Updated:** March 1, 2026
 
 ## 1. Scope and Authority
@@ -135,7 +135,7 @@ Unit handling policy is strict and deterministic.
 3. Implicit solver algebraic-loop handling.
 4. Automatic unit conversion ontology.
 
-## 11. Conformance Snapshot (March 1, 2026 - Post Phase 1 Pass)
+## 11. Conformance Snapshot (March 1, 2026)
 
 This section records current implementation alignment against the normative contract.
 
@@ -147,4 +147,4 @@ This section records current implementation alignment against the normative cont
 6. Unit checks enforce declared/first-write consistency and runtime mismatches throw: **partial** for Section 8 (compile-time inferential checks remain limited).
 7. Delay-mediated cycle acceptance is implemented via non-delay subgraph cycle policy: **partial** pending deeper delay-buffer semantics validation in complex mixed-order graphs.
 
-Remaining gaps define follow-up Phase 1 hardening and Phase 3 evidence work.
+Remaining gaps define follow-up hardening and expanded evidence work.

docs/validation-methodology.md

@@ -2,9 +2,9 @@
 
 ## Purpose
 
-This document defines the validation protocol used to support Phase 3
+This document defines the validation protocol used to support
 numerical claims for FluxGraph model integration behavior.
-Phase-level criteria mapping is documented in `docs/phase3-evidence.md`.
+This document is the authoritative criteria/source for numerical evidence.
 
 ## Systems Under Test
 

include/fluxgraph/viz/dot_emitter.hpp

@@ -15,7 +15,7 @@ struct DotEmitOptions {
 std::string emit_dot(const GraphSpec &spec, const DotEmitOptions &options = {});
 
 /// Return sorted transform types that are not part of the built-in set.
-std::vector<std::string> collect_extension_transform_types(const GraphSpec &spec);
+std::vector<std::string>
+collect_extension_transform_types(const GraphSpec &spec);
 
 } // namespace fluxgraph::viz
-

include/fluxgraph/viz/format.hpp

@@ -15,4 +15,3 @@ std::optional<OutputFormat> parse_output_format(const std::string &value);
 const char *output_format_name(OutputFormat format);
 
 } // namespace fluxgraph::viz
-

include/fluxgraph/viz/graphviz_renderer.hpp

@@ -23,4 +23,3 @@ std::string build_graphviz_command(const GraphvizRenderRequest &request);
 GraphvizRenderResult render_with_graphviz(const GraphvizRenderRequest &request);
 
 } // namespace fluxgraph::viz
-