perf: cache rendered CodeLens array to skip per-call object allocation

[github-actions]

🤖 This pull request was created by Repo Assist, an automated AI assistant.

Summary

provideCodeLenses was calling createCodeLenses on every invocation — even when the document content and workspace config were unchanged (cursor moves, scroll events, focus switches). Each call allocated a vscode.Range, vscode.Command, and vscode.CodeLens object for every tracked function.

This PR adds a codeLensCache that stores the rendered vscode.CodeLens[] array and returns it directly when both the document content (analysisKey) and workspace config (configKey) are unchanged.

---

Changes

src/providers/codeLensProvider.ts

• New codeLensCache = Map<string, vscode.CodeLens[]> keyed by "${analysisKey}#${configKey}" (uri#langId#version#folderUri).
• In provideCodeLenses: after resolving functions (from analysis cache or fresh parse), look up codeLensKey. Return cached lenses on hit; build and store on miss.
• LRU eviction: bounded to ANALYSIS_CACHE_MAX_SIZE (64) entries, consistent with the analysis cache it depends on.
• clearConfigCache(): also clears codeLensCache — threshold changes alter icon/text in every lens.
• clearAnalysisCache(): also clears codeLensCache — language mode switches invalidate rendered lenses.
• pruneAnalysisCacheForDocument(): also removes matching codeLensCache entries on document close.

---

When does this help?

VS Code calls provideCodeLenses on focus events, scroll events, and other triggers beyond text changes. With this cache:

• Same document version + same config → zero object allocation (pure cache hit).
• Document edited → analysisKey changes; fresh analysis + fresh render.
• Config changed → both caches are cleared; fresh render on next call.

For a file with 50 analysed functions, each provideCodeLenses call previously allocated 50 × (1 Range + 1 Command + 1 CodeLens) = 150 objects on every cursor move. After this PR: zero.

---

Test Status

npm run compile  ✅  (0 errors)
npm run lint     ✅  (0 warnings)
npm run test:unit  ✅  198 passing, 0 failing

Warning

Firewall blocked 1 domain

The following domain was blocked by the firewall during workflow execution:

• releaseassets.githubusercontent.com

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "releaseassets.githubusercontent.com"

See Network Configuration for more information.

Generated by 🌈 Repo Assist, see workflow run. Learn more.

Add this agentic workflows to your repo

To install this agentic workflow, run

gh aw add githubnext/agentics/workflows/repo-assist.md@1c6668b751c51af8571f01204ceffb19362e0f66

[codecov]

Codecov Report

❌ Patch coverage is 0% with 42 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (main@e3bdcc2). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
src/providers/codeLensProvider.ts	0.00%	42 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #474   +/-   ##
=======================================
  Coverage        ?   81.40%           
=======================================
  Files           ?       13           
  Lines           ?     4345           
  Branches        ?      440           
=======================================
  Hits            ?     3537           
  Misses          ?      807           
  Partials        ?        1           

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Copilot]

Pull request overview

This PR improves the VS Code extension’s runtime performance by caching the fully rendered vscode.CodeLens[] output so repeated provideCodeLenses invocations (triggered by non-text-change editor events) can avoid re-allocating Range/Command/CodeLens objects when inputs are unchanged.

Changes:

• Added an LRU-bounded codeLensCache keyed by analysisKey + workspace-folder configKey and return cached lenses on hit.
• Wired cache invalidation into existing config/analysis cache clearing, and pruned cached CodeLens entries on document close.
• Kept cache sizing consistent with the existing analysis cache limit (ANALYSIS_CACHE_MAX_SIZE).