docs: drop completed work from TODO.md

[bernardladenthin]

Summary

• TODO.md carried a Done (kept for history) section and two already-resolved items, which made the open list look three times longer than it is. 34 → 20 lines. Docs only.
• Both removed items were verified, not taken on trust:
  • SpotBugs effort=Max + threshold=Low — pom.xml carries <effort>Max</effort> and <threshold>Low</threshold>, and spotbugs:check runs in verify.
  • "No LogCaptor smoke test needed" — src/main/java has no org.slf4j usage at all, so it documents a problem this repo does not have.
• The Done (kept for history) block (12 finished items) goes too. Git history and crossrepostatus.md already hold that record; duplicating it here only ages.

The @VisibleForTesting entry is kept, but sharpened

It said "No usages currently", which reads as "nothing to do". StreamBuffer actually has 15 package-private methods that exist so tests can reach them — decideTrimExecution, shouldTrim, clampToMaxInt, decrementAvailableBytesBudget, calculateResultingChunks, the shouldSkipTrim* predicates, isAvailableBytesPositive, isMaxAllocSizeLessThanAvailable, shouldCheckEdgeCase, recordReadStatistics, shouldUpdateMaxObservedBytes, updateMaxObservedBytesIfNeeded.

None is annotated and Guava is not a dependency, so the entry now states the actual decision it needs: (a) add a project-local marker annotation — which puts a new public type into the API surface of a deliberately one-class library — or (b) record that the convention does not apply here. Either closes it; leaving it as a standing "audit" does not.

Test plan

• Diff is TODO.md only — no source, test or build file touched
• Both removed claims re-checked against the current tree rather than believed
• CI is green on this branch — docs-only, but let it run

Related issues / PRs

Part of a four-repo pass; siblings are the matching claude/todo-cleanup branches in java-llama.cpp, BitcoinAddressFinder and srcmorph. Each file's header now states the rule outright: finished items are deleted, not annotated.

Checklist

• I have read CONTRIBUTING.md and CODE_OF_CONDUCT.md
• My commits follow Conventional Commits
• No security-sensitive changes

---

Generated by Claude Code

[claude]

Minor wording issue: The text says "the five shouldSkipTrim*/should* predicates" but there are only four shouldSkipTrim* methods in the codebase:

• shouldSkipTrimDueToEdgeCase
• shouldSkipTrimDueToInvalidMaxBufferElements
• shouldSkipTrimDueToSmallBuffer
• shouldSkipTrimDueToSufficientBuffer

Suggest changing "five" to "four" for accuracy, or if the count was meant to include additional predicates, clarify which ones are grouped under "the five" vs. the separately-listed methods.

[claude]

Review Summary

This is a documentation-only PR that cleans up completed work from TODO.md. Overall, the changes are well-motivated and the claims have been verified against the current codebase.

Verified claims:

✅ SpotBugs effort=Max + threshold=Low: Confirmed in pom.xml; spotbugs:check runs during mvn verify
✅ No LogCaptor smoke test needed: Verified — no org.slf4j.* usage in src/main/java/
✅ 15 package-private test-accessible methods: Accurate count confirmed; all methods listed exist and are package-private

Improvements in clarity:

• New header explaining the convention (finished items deleted, not annotated) is helpful
• @VisibleForTesting audit expanded with concrete method list and two actionable decision paths (a/b)
• jqwik entry clarified as a standing constraint requiring re-check on dependency bumps
• Cross-repo items scoped more clearly

Code quality assessment:

• ✅ No security concerns (docs-only change)
• ✅ No breaking changes
• ✅ Follows the stated convention from CLAUDE.md
• ✅ Git history and crossrepostatus.md remain as the source of truth

Note: See inline comment regarding a minor wording issue with method count.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.88%. Comparing base (25c2cb7) to head (b3eb446).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##               main     #153   +/-   ##
=========================================
  Coverage     96.88%   96.88%           
  Complexity       94       94           
=========================================
  Files             1        1           
  Lines           257      257           
  Branches         33       33           
=========================================
  Hits            249      249           
  Misses            2        2           
  Partials          6        6           

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

[coveralls]

coverage: 99.222%. remained the same — claude/todo-cleanup into main