fix(gateway): use OGG/Opus for Matrix voice replies

[liuhao1024]

What does this PR do?

Matrix MSC3245 voice messages require Opus audio format, but both the _send_voice_reply path in gateway/run.py and the want_opus check in tools/tts_tool.py only triggered Opus conversion for Telegram. Matrix voice replies were sent as MP3/WAV, which Element X silently refuses to play in the voice bubble.

Related Issue

Fixes #58715

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)

Changes Made

• gateway/run.py: Add Platform.MATRIX to the _voice_platforms tuple so _send_voice_reply generates .ogg files for Matrix sessions
• tools/tts_tool.py: Add "matrix" to the want_opus condition so the TTS tool triggers ffmpeg Opus conversion for Matrix platform sessions
• tests/tools/test_tts_opus_routing.py: Add regression test verifying Edge TTS converts to Opus when HERMES_SESSION_PLATFORM=matrix

How to Test

1. Configure a Matrix gateway with any TTS provider (e.g., Edge TTS or piper)
2. In a room, send /voice tts to enable voice replies
3. Send a text message and wait for the voice reply
4. The voice bubble should now play correctly in Element X iOS — previously it showed the bubble with correct duration but playback did nothing
5. Run pytest tests/tools/test_tts_opus_routing.py -v — all 3 tests should pass, including the new test_edge_matrix_converts_to_opus_voice

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/tools/test_tts_opus_routing.py -v and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: macOS 15 (Apple Silicon)

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

[alt-glitch]

This was generated by AI during triage.

Duplicate of #14900, an earlier open PR fixing the same bug (#14841) via the same mechanism: adding matrix to the .ogg extension logic in gateway/run.py and to the want_opus check in tools/tts_tool.py. #14900 is the canonical fix. Also related: #54488 (a competing approach that transcodes MP3->Ogg/Opus after generation and adds native MSC3245 metadata). A maintainer should pick between the predicate-extension approach (#14900/this) and the transcode approach (#54488).

[liuhao1024]

Closing as duplicate of #14900 (Kailigithub) — same Matrix OGG/Opus voice reply fix in gateway/run.py + tools/tts_tool.py, created 2+ months earlier.

This PR's regression test (test_edge_matrix_converts_to_opus_voice) may be worth porting to #14900 if the maintainer requests it.