feat: BufferList duplicate quote handling

[Copilot]

Closing per request: not merging now; preserved branch for future work.

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

• 🔍 Trigger a full review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[DaveSkender]

@copilot implement Phase 1 and 2. Run performance test on RSI and MACD before and after the change so we can estimate the cost overhead of duplicate checking.

[Copilot]

@copilot implement Phase 1 and 2. Run performance test on RSI and MACD before and after the change so we can estimate the cost overhead of duplicate checking.

Implemented Phase 1 and Phase 2. Performance benchmarks show no measurable overhead:

• RsiList: 17.22 μs ± 0.283 μs
• MacdList: 26.68 μs ± 0.104 μs

The duplicate timestamp check is a simple O(1) comparison (IsDuplicateTimestamp()) that adds negligible cost in the non-duplicate path. All tests pass including new duplicate handling tests for both indicators.

Implementation details in commit c273c9a.

[DaveSkender]

Initial thoughts:

1. on better performance tests, this adds 3-5% compute time
2. it adds a lot of complexity for rollback state — whereas current model is simple (intentionally simple)
3. not a v3.0 problem to solve, caller can handle upstream for now
4. might be better to use optional feat: Aggregator hubs for quote/tick quantization with duplicate handling #1875