Add Qwen3-VL and WebRTC dictation services Add Bazel targets for the CUDA-backed Qwen3-VL server and a local WebRTC faster-whisper dictation service. Co-authored-by: Copilot <[email protected]> Copilot-Session: e3d8cb06-6c95-4ae0-9757-651d3796ab00

aiortc==1.15.0
fastapi==0.141.1
faster-whisper==1.2.1
huggingface-hub==1.27.0
httpx==0.28.1
nvidia-cublas-cu12==12.9.2.10
nvidia-cudnn-cu12==9.24.0.43
setuptools==84.0.0
uvicorn==0.52.3