extract shared transcript parser and similarity matching helpers

- New agents/transcript.rs: shared JSONL parsing for enrich, fact_mine,
  and knowledge (was 3 separate implementations, ~150 lines duplicated)
- New best_match() and section_children() helpers in neuro/rewrite.rs
  (was duplicated find-best-by-similarity loop + section collection)
- Net -153 lines

poc-memory/src/agents/mod.rs

@@ -13,7 +13,9 @@
 //   fact_mine    — fact extraction from transcripts
 //   digest       — episodic digest generation (daily/weekly/monthly)
 //   daemon       — background job scheduler
+//   transcript   — shared JSONL transcript parsing
 
+pub mod transcript;
 pub mod llm;
 pub mod prompts;
 pub mod audit;