test(sessions): add Session/Memory/Summary replay consistency harness

[quan020406]

Related Issues

Resolved #89

Summary

Adds a replay consistency harness for Session, Memory, and Summary behavior across InMemory and SQLite backends.

The harness loads 10 public replay cases, applies identical operation sequences to each backend, normalizes non-business differences, compares SDK-observable state, and writes a structured diff report.

Changes

• Add 10 public replay cases covering single-turn, multi-turn, tool calls, state updates, memory store/search, summary generation/update, summary truncation, duplicate write recovery, and wrong-session isolation.
• Add tests/sessions/test_replay_consistency.py for InMemory vs SQLite replay comparison.
• Add mutation checks for injected event, state, memory, and summary inconsistencies.
• Generate session_memory_summary_diff_report.json with case id, backend pair, field path, session id, event index or summary id, and compared values.
• Add a 150-300 word design note explaining normalization, summary comparison, allowed differences, and backend integration.

Results

• pytest tests\sessions\test_replay_consistency.py -q
• Result: 4 passed, 1 skipped
• Redis integration is gated by TRPC_AGENT_REPLAY_REDIS_URL and skipped when unset.
• Normal replay cases report 0 diffs.
• Injected inconsistency checks detect all configured mutations.
• Summary lost, summary overwrite error, and summary wrong-session ownership are all detected.

Notes

The current SDK does not expose a dedicated persisted summary.version field. This harness interprets summary version as observable revision state, validated through generation order, replacement behavior, persisted summary text, session ownership, timestamp validity, and coverage/projection consistency.

Backward Compatibility

No production SDK API changes. All changes are limited to tests, fixtures, report output, and design documentation.

RELEASE NOTES: NONE

[github-actions]

CLA Assistant Lite bot All contributors have signed the CLA ✍️ ✅

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@4336564). Learn more about missing BASE report.

Additional details and impacted files

@@            Coverage Diff             @@
##             main        #158   +/-   ##
==========================================
  Coverage        ?   87.55867%           
==========================================
  Files           ?         467           
  Lines           ?       44103           
  Branches        ?           0           
==========================================
  Hits            ?       38616           
  Misses          ?        5487           
  Partials        ?           0           

☔ 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.

[quan020406]

I have read the CLA Document and I hereby sign the CLA