feat(stt): mode texte + sélection de modèle (client→serveur par requête) (#7)

Serveur : /v1/ask accepte un model optionnel (validé contre STT_ALLOWED_MODELS),
nouvel endpoint GET /v1/models (défaut + alias autorisés). Pas de switch global.

Client : 'stt --text' (chat texte simple sans micro/HUD), '--model hermes|claude|qwen|opus'
(noms courts → alias LiteLLM), commandes /model et /models en mode texte. Le modèle
choisi est envoyé au serveur à chaque requête (voix comme texte).


Claude-Session: https://claude.ai/code/session_013FmcxGsyXZXogiAHQLjnZT

Co-authored-by: Claude <noreply@anthropic.com>

stt/server/stt_server/app.py

@@ -13,16 +13,19 @@ from pydantic import BaseModel
 
 from stt_server import __version__
 from stt_server.brain import ask as brain_ask
+from stt_server.config import settings
 
 app = FastAPI(title="STT-server", version=__version__)
 
 
 class AskRequest(BaseModel):
     text: str
+    model: str | None = None   # alias LiteLLM ; défaut serveur si absent
 
 
 class AskReply(BaseModel):
     reply: str
+    model: str
 
 
 @app.get("/healthz")
@@ -30,16 +33,27 @@ async def healthz() -> dict:
     return {"status": "ok", "version": __version__}
 
 
+@app.get("/v1/models")
+async def v1_models() -> dict:
+    return {"default": settings.model, "available": settings.allowed_models}
+
+
 @app.post("/v1/ask", response_model=AskReply)
 async def v1_ask(req: AskRequest) -> AskReply:
     text = req.text.strip()
     if not text:
         raise HTTPException(status_code=400, detail="text vide")
+    model = req.model or settings.model
+    if model not in settings.allowed_models:
+        raise HTTPException(
+            status_code=400,
+            detail=f"modèle '{model}' non autorisé ; dispo : {settings.allowed_models}",
+        )
     try:
-        reply = await brain_ask(text)
+        reply = await brain_ask(text, model)
     except httpx.HTTPError as e:
         raise HTTPException(status_code=502, detail=f"upstream LiteLLM : {e}") from e
-    return AskReply(reply=reply)
+    return AskReply(reply=reply, model=model)
 
 
 def run() -> None: