feat: Auto-save voice recordings for model training

- Add /api/recordings endpoint with full CRUD operations
- Create voice_recordings SQLite table for metadata
- Save audio files to server/recordings/ as .webm
- Store transcription, duration, microphone name, file size
- Auto-save on each Whisper recording completion

server/db/migrations.ts

@@ -85,6 +85,20 @@ export function runMigrations(db: Database) {
     )
   `)
 
+  // Voice recordings table (for training custom speech models)
+  db.run(`
+    CREATE TABLE IF NOT EXISTS voice_recordings (
+      id INTEGER PRIMARY KEY AUTOINCREMENT,
+      filename TEXT NOT NULL,
+      transcription TEXT,
+      duration_ms INTEGER,
+      microphone TEXT,
+      sample_rate INTEGER,
+      file_size INTEGER,
+      created_at TEXT DEFAULT CURRENT_TIMESTAMP
+    )
+  `)
+
   // Run column migrations for existing tables
   runColumnMigrations(db)
 }