feat: extract Python method body references for dependency graph

[Zir0-93]

Summary

• daemon.js: Add collectBodyReferences() that walks the suite AST to find Call nodes and resolve their callees via resolveTypeFromRaw(). Skips nested Function/Class nodes. Results attached to both extractMethod() and extractFunction() return values.
• PythonMethodModel: Add bodyReferences field (List<PythonTypeRefModel>)
• PythonModelAssembler: Wire body refs as SimpleTypeReference entries in both buildMethodComponent() and buildFunctionComponent()

This dramatically increases relationship density for Python projects (~27 rels/component → potentially 100+), which was the root cause of Python's 0% match rate in the existential test.

Test plan

• 5 new unit tests in PythonMethodBodyRefsTest — all passing
• Full Python test suite (112 tests) passes
• Verify on real Python repos (sqlmodel, werkzeug, jinja) that rel density increases

🤖 Generated with Claude Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 828ffd45bb

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Avoid resolving instance calls as module functions

When the callee is a dotted expression, resolving the entire text through resolveTypeFromRaw makes ordinary instance calls look like module-level symbols whenever the receiver name matches a module in the package. In the added fixture, dao = UserDao(); dao.find_by_id(...) is emitted by the daemon as a targetUniqueName of src.dao.find_by_id (not the UserDao.find_by_id(...) method component), so after classification the method gains a bogus external dependency. This will pollute dependency graphs for common variable names like repo.save() or service.call() that happen to match sibling modules; dotted callees should only be resolved this way when the base is a known imported module/alias, not a local variable.

Useful? React with 👍 / 👎.

[codecov]

Codecov Report

❌ Patch coverage is 38.46154% with 8 lines in your changes missing coverage. Please review.
✅ Project coverage is 71.77%. Comparing base (452cd06) to head (58868a6).

Files with missing lines	Patch %	Lines
.../clarpse/compiler/python/PythonModelAssembler.java	33.33%	4 Missing and 4 partials ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##             master     #147      +/-   ##
============================================
- Coverage     71.87%   71.77%   -0.11%     
- Complexity      903      904       +1     
============================================
  Files            65       65              
  Lines          4093     4106      +13     
  Branches        699      705       +6     
============================================
+ Hits           2942     2947       +5     
- Misses          723      727       +4     
- Partials        428      432       +4     

☔ View full report in Codecov by Sentry.
📢 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.