fix: voice input in Electron - local Whisper STT + network error handling

- Add local Whisper STT via Transformers.js with Web Worker (no UI freeze)
- Default sttProvider to 'local' in Electron (browser STT unavailable)
- Fix infinite toast loop: stop auto-restart on network errors
- Add retry limit with exponential backoff for transient STT errors
- Append voice transcript to input field (append-inline), not replace
- Add model catalog with download/load button in Voice Settings

packages/ui/src/lib/i18n/messages/es.settings.ts

@@ -1301,6 +1301,14 @@ export const settingsDict = {
   "settings.voice.page.provider.custom": "Personalizado",
   "settings.voice.page.provider.say": "Decir",
   "settings.voice.page.provider.server": "Servidor",
+  "settings.voice.page.provider.wasm": "Local",
+  "settings.voice.page.stt.wasmModel": "Modelo Whisper",
+  "settings.voice.page.stt.wasmLoaded": "Modelo cargado",
+  "settings.voice.page.stt.wasmDownloading": "Descargando modelo...",
+  "settings.voice.page.stt.wasmLoading": "Cargando modelo...",
+  "settings.voice.page.stt.wasmNotLoaded": "El modelo se descargará en el primer uso de voz",
+  "settings.voice.page.stt.wasmDownload": "Descargar",
+  "settings.voice.page.stt.wasmRetry": "Reintentar",
   "settings.voice.page.tooltip.browser": "Gratuito, offline, soporte limitado en móvil.",
   "settings.voice.page.tooltip.openai": "Calidad alta, compatible con móviles, requiere clave de API.",
   "settings.voice.page.tooltip.custom": "Servidor compatible con OpenAI (por ejemplo Kokoro).",