fix(recall): remove score/source suffixes from visible recall text

[Disaster-Terminator]

Summary

Remove retrieval score/source suffixes from rendered recall text while preserving the structured retrieval details schema.

What changes

• remove the (score%, sources) suffix from memory_recall text output in src/tools.ts
• remove the same suffix from auto-recall injected lines in index.ts
• keep details.memories[*].score and details.memories[*].sources unchanged
• add regression coverage for:
  • exact memory_recall rendered text output
  • multi-line memory_recall rendering with an expanded result set
  • before_agent_start auto-recall injection

What does not change

• no schema changes
• no retriever/ranking behavior changes
• no recall-engine debug/logging changes
• CLI memory-pro search output remains unchanged in this PR
• no unrelated refactors

Why

Rendered recall text is prompt-visible context. Structured retrieval details remain available for programmatic use, but appending score/source metadata to visible recall text adds prompt noise and unnecessary formatting clutter.

Validation

• node --test test/recall-text-cleanup.test.mjs ✅
• npm test ✅

[win4r]

@claude review

[win4r]

@codex review

[claude]

Claude Code is working…

I'll analyze this and get back to you.

View job run

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Already looking forward to the next diff.

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".