Feat/ai contract completion assistance

[Power70]

Description

Adds starforge complete, an intelligent offline completion assistant for
Soroban contracts. It suggests context-aware code completions, generates
accurate boilerplate, fills in function stubs, suggests imports, and infers
types. The engine is rule-based and dependency-free (std-only), so it runs
instantly with no network access or model download.

Closes #508

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Changes Made

• Add src/utils/completion.rs — std-only completion engine: source analysis, context-aware suggestions, boilerplate generation, function-stub completion, import inference, and type inference.
• Add src/commands/complete.rs — starforge complete CLI with suggest, boilerplate, stub, imports, and infer subcommands (human + --json output).
• Register the new modules and wire the Complete command into src/main.rs, src/commands/mod.rs, and src/utils/mod.rs.
• Add tests/completion_assistant.rs — CLI integration tests.
• Incidental build fix: remove stray match-arm lines left inside enum Commands, wire the previously-unregistered Migrate variant into both dispatch matches, and drop a duplicate pub mod simulate; — master did not compile without these.

Testing

How has this been tested?

The std-only engine and the stub-splicing logic were compiled and run
standalone with rustc --test (the full crate could not be built in the dev
sandbox — see Additional Context). CLI behaviour is covered by integration
tests that shell out to the built binary.

Reproduce:

cargo test --lib completion
cargo test --test completion_assistant

• Unit tests added/updated
• Integration tests added/updated
• Manual testing performed

Test Coverage

Describe what scenarios have been tested:

• Happy path: boilerplate generation for each kind, suggestions on a partial contract, import/type inference, stub --write applying generated bodies.
• Edge cases: empty file (scaffold suggestion), imported-but-unused symbols, un-annotated let bindings incl. mut, brace balance/indentation preserved when splicing stub bodies.
• Error handling: missing file returns a non-zero exit; unknown boilerplate kind is rejected with the list of valid kinds.

Code Quality Checklist

• My code follows the style guidelines of this project (cargo fmt)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings (cargo clippy -- -D warnings)
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• The CI checks pass (format, clippy, tests)

cargo fmt, cargo clippy, and the full test suite could not be run locally
(offline sandbox); please confirm these in CI.

Breaking Changes

• This PR introduces breaking changes

None — this adds a new, self-contained command and does not alter existing behaviour.

Documentation

• README.md updated
• DEVELOPER_GUIDE.md updated (if applicable)
• API_REFERENCE.md updated (if applicable)
• No documentation changes needed

Module- and command-level rustdoc is included; usage is discoverable via
starforge complete --help.

Screenshots (if applicable)

N/A — CLI feature.

Additional Context

The full crate could not be compiled in the development sandbox: the crates.io
registry was unreachable and tokio's transitive dependencies were not cached.
Logic was therefore validated by compiling the std-only portions standalone
with rustc --test (23 engine unit tests pass; stub-splicing verified). The
thin CLI layer follows existing command patterns (migrate.rs, gas.rs,
lint.rs). Please run the checks below in CI.

---

Note: Make sure all tests pass locally before submitting:

cargo test
cargo fmt --all
cargo clippy -- -D warnings

[drips-wave]

@Power70 Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits