Add LM Model selector (0.6B/1.7B/4B) with auto-download

- Add lmModel parameter through full chain (types, API, routes, service)
- Add LM Model dropdown in Advanced Settings (defaults to 0.6B)
- Pass lmModel and lmBackend to format/enhance endpoint
- Update format_sample.py to accept --lm-model and --lm-backend args
- Auto-download model from HuggingFace if not present locally
- Persist model selection in localStorage
- Improve format route error logging with exit code and stdout/stderr

Fixes #9

App.tsx

@@ -722,6 +722,7 @@ export default function App() {
         lmTopP: params.lmTopP,
         lmNegativePrompt: params.lmNegativePrompt,
         lmBackend: params.lmBackend,
+        lmModel: params.lmModel,
         referenceAudioUrl: params.referenceAudioUrl,
         sourceAudioUrl: params.sourceAudioUrl,
         referenceAudioTitle: params.referenceAudioTitle,