refactor: address SOLID cleanup debt

[MichaelFisher1997]

Summary

• replace hardcoded non-cube mesher calls with a render-shape strategy table
• remove unused segregated worldgen interface wrappers and re-exports
• clean up job-system test cleanup state, document the stateless lighting interface token, and lock createShader consistently

Verification

• nix develop --command zig fmt src/ modules/
• nix develop --command zig build test

Closes #805
Closes #795
Closes #794

[github-actions]

📋 Summary

This PR addresses three linked issues:

• Closes [SOLID] [m3] Misc cleanup: test-only globals, fake anyopaque, createShader mutex (covers m4, m6) #805: cleanup_count test state is now passed per-test instead of file-scoped; interface_token is const with documentation; createShader now holds ctx.mutex.
• Closes [SOLID] [M7] Replace hardcoded mesher dispatch with strategy table #795: Hardcoded mesher calls replaced with SHAPE_MESHERS strategy table.
• Closes [SOLID] [C5] Consume or delete dead segregated worldgen interfaces #794: Unused segregated worldgen interface wrappers (IChunkGenerator, ILODHeightmapGenerator, IGeneratorInfoProvider, ICacheRecenterable) and their re-exports are removed.

Scope is a behavior-preserving SOLID cleanup across job-system tests, meshing dispatch, worldgen API surface, and RHI locking. Code is well formatted, tests pass, and no consumers of the removed interfaces remain.

🔴 Critical Issues

None identified.

⚠️ High Priority Issues

None identified.

💡 Medium Priority Issues

None identified.

ℹ️ Low Priority Suggestions

[LOW] modules/world-meshing/src/chunk_mesh.zig:195-200 - Strategy table shape field is unused for dispatch
Confidence: Medium
Description: SHAPE_MESHERS carries a shape: RenderShape field, but the loop suppresses it with _ = entry.shape; and dispatches purely by table position. The linked issue #795 requested a table "keyed on render_shape"; the current implementation has the metadata but does not use it to route shapes.
Impact: The refactor is behaviorally correct, but the unused field is a mild maintainability smell and the table does not yet fully satisfy the "keyed dispatch" intent.
Suggested Fix: Either remove the shape field until it is needed, or add a std.debug.assert(entry.shape == expected_shape) / comptime validation so the metadata stays honest. A future step could refactor meshers to accept a shape filter and dispatch via the table key.

📊 SOLID Principles Score

Principle	Score	Notes
Single Responsibility	9	Each commit area has one clear concern (dispatch, API cleanup, test hygiene, locking).
Open/Closed	8	Strategy table makes adding meshers easier, though dispatch is positional rather than shape-keyed.
Liskov Substitution	9	No subtype changes; removal of unused wrappers tightens contracts.
Interface Segregation	9	Dead segregated interfaces removed, reducing surface area.
Dependency Inversion	8	Meshing dispatch depends on an abstract table rather than concrete call order.
Average	8.6

🎯 Final Assessment

Overall Confidence Score: 95%

Confidence Breakdown:

• Code Quality: 95%
• Completeness: 95%
• Risk Level: 90% (low risk)
• Test Coverage: 95%

Merge Readiness:

• All critical issues resolved
• SOLID average score >= 6.0
• Overall confidence >= 60%
• No security concerns
• Tests present and passing

Verdict:

MERGE

Clean, behavior-preserving refactor with passing tests and addressed linked issues.

{
  "reviewed_sha": "8b7dc789c0578de0a9b8eec97b50741d76cbb38a",
  "critical_issues": 0,
  "high_priority_issues": 0,
  "medium_priority_issues": 0,
  "overall_confidence_score": 95,
  "recommendation": "MERGE"
}

opencode session  |  github run