test(cargo-gamma): keep runtime fixtures offline

[martin-kolinek]

Problem

Build 40731154 introduced two cargo-gamma CLI tests that create temporary workspaces pointing at the repository's real cargo-gamma-rt manifest. Before cargo-gamma can redirect that dependency, cargo metadata resolves its workspace-inherited loom dependency and contacts index.crates.io. The official pipeline enforces network isolation, so both tests fail; the missing clippy SARIF and Rust evidence directories are downstream symptoms.

Fix

Use a dependency-free local cargo-gamma-rt stub as the fixture's pre-redirect dependency. The stub declares embedding but deliberately omits gamma_rt::embedded, preserving the test's proof: it succeeds only if cargo-gamma redirects to the real vendored runtime and retains the feature.

Verification

• Both affected tests pass on Windows with CARGO_NET_OFFLINE=true
• Both affected tests pass on Linux with CARGO_NET_OFFLINE=true
• Full cargo-gamma-lib nextest suite passes (1,948 tests)

[Copilot]

🟢 Approval recommended

The changes are localized to test fixtures and directly address the stated offline-CI failure mode without introducing apparent behavioral risk outside the affected tests.

Pull request overview

This PR fixes cargo-gamma CLI integration tests failing under network-isolated CI by ensuring the runtime dependency used in test fixtures does not trigger registry access before cargo-gamma performs its dependency redirection.

Changes:

• Added a local, dependency-free cargo-gamma-rt “runtime-stub” fixture generator for CLI tests.
• Updated the two affected tests to depend on the local stub runtime instead of the repository’s real runtime crate path, preventing cargo metadata from touching crates.io before redirection.

File summaries

File	Description
crates/cargo-gamma-lib/tests/cli.rs	Introduces a local runtime stub and rewires two CLI tests to use it, keeping fixtures offline until cargo-gamma redirects dependencies.

Review details

• Files reviewed: 1/1 changed files
• Comments generated: 0
• Review effort level: Lite

---

💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Codecov Comments Bot (codecov-commenter)]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.1%. Comparing base (26e5c60) to head (f12852a).
⚠️ Report is 5 commits behind head on main.

❌ Your project status has failed because the head coverage (93.1%) is below the target coverage (100.0%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files

@@           Coverage Diff            @@
##            main    #171      +/-   ##
========================================
- Coverage   97.5%   93.1%    -4.5%     
========================================
  Files        300     154     -146     
  Lines      68538   44178   -24360     
========================================
- Hits       66876   41155   -25721     
- Misses      1662    3023    +1361     

Flag	Coverage Δ
linux	93.2% <ø> (-4.4%)	⬇️
linux-arm	93.2% <ø> (-4.4%)	⬇️
scheduled	?
windows	93.8% <ø> (-3.9%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.