sync: bring the Space up to the current GitHub tree

fix(web): raise /api proxy timeout to 300s — CPU LLM generation exceeds Next's 30s default on non-streaming endpoints

feat(voice): NotebookLM-style voice UX

HardwareGate: detect the visitor's real device client-side

ModelFit: clarify server-vs-device on hosted Space + enable wiki

Deploy Auralynq RAG (Llama-3.3-70B via HF Inference Providers)