Add /v1/models endpoint for voice discovery

- Modified openedai.py to allow speech.py to define custom /v1/models
- Endpoint returns comprehensive model info including:
  * All available voices per model
  * Voice count
  * Engine name (piper, xtts, silero, kokoro)
  * Sample rate
  * Description
- Supports all 4 TTS engines:
  * tts-1 (Piper): 55 voices @ 22050 Hz
  * tts-1-hd (XTTS): 8 voices @ 24000 Hz
  * tts-1-silero (Silero): 148 voices @ 48000 Hz
  * tts-1-kokoro (Kokoro): 34 voices @ 24000 Hz

This enables frontends to dynamically discover available voices
and their supported models without hardcoding voice lists.

🦝 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

openedai.py

@@ -145,9 +145,11 @@ class OpenAIStub(FastAPI):
         async def health():
             return {"status": "ok" if self.models else "unk" }
 
-        @self.get("/v1/models")
-        async def get_model_list():
-            return self.model_list()
+        # NOTE: /v1/models endpoint is defined in speech.py for custom voice listing
+        # If you need the default behavior, uncomment these lines:
+        # @self.get("/v1/models")
+        # async def get_model_list():
+        #     return self.model_list()
 
         @self.get("/v1/models/{model}")
         async def get_model_info(model_id: str):