feat(voice): streaming TTS end-to-end with diagnostics, Aidoru theming, visible TTFA

fix(air-gaps): strip fade envelope from worklet, batch audio chunks, add diagnostics

Queue-clear on active switch + JS-side resampling

Resample 24kHz->native in worklet + diagnostic logging

Streaming TTS: worklet playback + 30ms crossfade

Add audio-worklet.js