## Summary
- add server-backed speech capabilities and transcription endpoints plus
UI settings for speech configuration
- add push-to-talk prompt voice input with microphone controls,
transcription insertion, and browser capability gating
- keep prompt controls aligned by restoring right-side nav placement and
moving the mic beside the expand control
Add generic /api/storage config/state endpoints with merge-patch, migrate legacy YAML/JSON layout, and update UI/server to read and write owner-scoped settings. Replace config SSE events and drop /api/config routes.
Publish bleeding-edge builds from dev to GitHub prereleases and npm dist-tag 'dev'. Dev builds poll GitHub prereleases and surface update availability via /api/meta for UI notifications.
Default to HTTPS with optional loopback HTTP, generate/rotate self-signed certs via node-forge, and surface Local/Remote connection URLs. Update /api/meta schema, UI remote access overlay, and desktop shells to follow the new startup output.
Prevent desktop wrappers from SIGKILLing the CLI during shutdown, which could orphan OpenCode workspace processes. Shut down workspaces earlier/in parallel and increase the quit grace period.
Adds cookie-based login with a bootstrap token flow for desktop apps, secures OpenCode instance traffic with per-instance Basic auth, and updates UI/plugin clients to use credentials.
