Add test for AudioSegmentProcessor long segment splitting

[ysdede]

🎯 What:
Added a new test case to src/lib/audio/AudioSegmentProcessor.test.ts to verify the "Proactive Segment Splitting" logic in AudioSegmentProcessor.

📊 Coverage:
The test simulates a continuous high-energy audio stream exceeding the configured maxSegmentDuration (0.5s for the test) and asserts:

1. A segment is emitted at the expected duration.
2. The processor transitions correctly to a new speech segment without dropping state.
3. Timestamps (startTime, endTime) are consistent.

✨ Result:
Improved test coverage ensures that long utterances are split correctly, preventing potential latency or memory issues in the VAD pipeline.

---

PR created automatically by Jules for task 4745565586748818318 started by @ysdede

[ysdede]

Review Summary (Post r1→master merge)

Status: ✅ MERGEABLE

Changes: Adds a focused test for AudioSegmentProcessor long segment splitting behavior. Tests that segments exceeding maxSegmentDuration are proactively split while maintaining speech state.

Assessment: Good targeted test. Validates the proactive splitting logic which is important for the v4 pipeline. The test configures a short max duration (500ms) and verifies:

• Split triggers at the right time
• Segment duration is bounded
• Processor remains in speech state after split

Note: PR #69 also adds tests for AudioSegmentProcessor (more comprehensive). Check for overlap — #69 likely supersedes this. If both merge, ensure no duplicate test names.

Recommendation: ✅ Merge (but merge #69 after and check for conflicts/overlap)

[google-labs-jules]

Review Summary (Post r1→master merge)

Status: ✅ MERGEABLE

Changes: Adds a focused test for AudioSegmentProcessor long segment splitting behavior. Tests that segments exceeding maxSegmentDuration are proactively split while maintaining speech state.

Assessment: Good targeted test. Validates the proactive splitting logic which is important for the v4 pipeline. The test configures a short max duration (500ms) and verifies:

• Split triggers at the right time
• Segment duration is bounded
• Processor remains in speech state after split

Note: PR #69 also adds tests for AudioSegmentProcessor (more comprehensive). Check for overlap — #69 likely supersedes this. If both merge, ensure no duplicate test names.

Recommendation: ✅ Merge (but merge #69 after and check for conflicts/overlap)

Thank you for the review. I acknowledge the note about potential overlap with PR #69.