test(audio): improve test coverage for player and tts_engine to 96%+

[Orinks]

Summary

Improves test coverage for the audio module, closing #8.

Coverage Results

Module	Before	After
player.py	66%	96%
tts_engine.py	63%	97%
Total	65%	96%

Tests Added (36 new tests)

player.py tests:

• BASS audio system initialization and cleanup
• BASS init failure propagation
• Fallback playback via playsound3 threading
• Fallback ImportError and general exception handling
• Volume updates on active sound_lib streams
• is_playing exception handling
• stop() exception recovery (stream set to None)
• BASS_Free error handling during cleanup

tts_engine.py tests:

• pyttsx3 initialization success and failure paths
• Speech synthesis error handling
• Voice listing exception handling
• Voice selection by ID (not just name)
• Voice not found case
• Voice selection exception handling
• Rate setter with engine property update
• Cleanup with engine.stop() and exception suppression
• Natural time format edge cases (quarter-to, non-standard minutes)
• speak_time with all options (style + date)

Remaining Uncovered Lines

• player.py lines 24-25, 28-29: Module-level import branches for sound_lib (can't be hit without module reload)
• tts_engine.py lines 22-24, 182: Module-level pyttsx3 import and list_voices exception log line

Closes #8

[Orinks]

Superseded by PR #10 which was merged with the same coverage improvements.