Update docs for cloud-first UX and shared captions

- README: document cloud-first quick start, shared captions workflow
  (create room, join via share code, share existing room), and
  self-hosting option
- README: update default remote.mode from local to byok in config table
- CLAUDE.md: reflect cloud-first default, settings gating, and shared
  captions features

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

CLAUDE.md

@@ -11,9 +11,11 @@ Local Transcription is a cross-platform desktop application for real-time speech
 **Key Features:**
 - Cross-platform desktop app (Windows, macOS, Linux) via Tauri v2 + Svelte 5
 - Headless Python backend with FastAPI control API
-- Dual transcription modes: local Whisper or cloud Deepgram (managed/BYOK)
+- Cloud-first: defaults to Deepgram (BYOK) transcription; local Whisper also supported
+- Settings UI hides local-only options (model, VAD, timing) when in cloud mode
+- Start button gated on API key / login — shows guidance if not configured
+- Shared Captions: create rooms, share via codes, join with one click (hosted at caption.shadowdao.com)
 - Built-in web server for OBS browser source at `http://localhost:8080`
-- Optional multi-user sync via Node.js server
 - CUDA, MPS (Apple Silicon), and CPU support
 - Auto-updates, custom fonts, configurable colors