perf(voice): run STT+TTS on the GPU by default with CPU fallback

Both voice backends defaulted to CPU. Fix the "CUDA unavailable" gaps so
everything that benefits from the RTX 5050 uses it:

- MeloTTS venv had CPU-only torch (2.12.0+cpu) -> installed Blackwell-capable
  torch/torchaudio 2.11.0+cu128 (sm_120 verified with a real GPU matmul).
- faster-whisper CUDA loaded but transcribe() died with "libcublas.so.12 not
  found": installed nvidia-cublas-cu12 + nvidia-cudnn-cu12 into the whisper
  venv and inject those nvidia/*/lib dirs into the worker's LD_LIBRARY_PATH at
  spawn (the loader only honours it at exec).
- WSAI_WHISPER_DEVICE / WSAI_MELO_DEVICE now default to "auto": pick CUDA when
  present, else CPU, and each worker falls back to CPU if a CUDA load fails so
  the voice loop never dies on a GPU-less host.

Verified end-to-end through the real backend classes: both workers report
"ready on cuda"; steady-state STT ~170ms (was ~1350ms CPU), TTS ~4s first call
vs ~23s CPU. All 12 tests pass.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

wsai/backends/whisper_worker.py

@@ -39,17 +39,42 @@ def _log(*a):
 
 def main() -> None:
     model_name = os.environ.get("WSAI_WHISPER_MODEL", "small")
-    device = os.environ.get("WSAI_WHISPER_DEVICE", "cpu")  # "cpu" | "cuda"
-    # int8 on CPU keeps a small model fast; float16 is the usual CUDA choice.
-    compute = os.environ.get(
-        "WSAI_WHISPER_COMPUTE", "int8" if device == "cpu" else "float16"
-    )
+    requested = os.environ.get("WSAI_WHISPER_DEVICE", "auto")  # cpu | cuda | auto
     default_lang = os.environ.get("WSAI_WHISPER_LANGUAGE", "ko") or None
 
-    t0 = time.monotonic()
     from faster_whisper import WhisperModel  # heavy import; only in whisper venv
 
-    model = WhisperModel(model_name, device=device, compute_type=compute)
+    def _has_cuda() -> bool:
+        try:
+            import ctranslate2
+
+            return ctranslate2.get_cuda_device_count() > 0
+        except Exception:
+            return False
+
+    device = requested
+    if requested == "auto":
+        device = "cuda" if _has_cuda() else "cpu"
+
+    def _compute_for(dev: str) -> str:
+        # int8 on CPU keeps a small model fast; float16 is the usual CUDA choice.
+        return os.environ.get(
+            "WSAI_WHISPER_COMPUTE", "int8" if dev == "cpu" else "float16"
+        )
+
+    t0 = time.monotonic()
+    try:
+        model = WhisperModel(model_name, device=device, compute_type=_compute_for(device))
+    except Exception as exc:
+        # CUDA picked but unusable (missing libs, OOM): fall back to CPU rather
+        # than leaving the whole voice loop dead.
+        if device == "cuda":
+            _log(f"[whisper_worker] CUDA load failed ({exc}); falling back to CPU")
+            device = "cpu"
+            model = WhisperModel(model_name, device=device, compute_type=_compute_for(device))
+        else:
+            raise
+    compute = _compute_for(device)
     load_ms = int((time.monotonic() - t0) * 1000)
     _emit({"ready": True, "ms": load_ms, "device": device, "model": model_name})
     _log(f"[whisper_worker] {model_name} ready in {load_ms} ms on {device}/{compute}")