Enable cudnn.benchmark to fix MIOpen workspace=0 on convolutions

Timing showed s3gen.inference (HiFiGAN vocoder) taking 22s and ref audio
processing ~18s - both dominated by Conv1d ops hitting MIOpen fallback.

With benchmark=False (default), PyTorch passes ptr=0 size=0 workspace to
MIOpen causing GemmFwdRest to fail and fall back to a slow path every call.
With benchmark=True, PyTorch evaluates convolution algorithms with proper
workspace allocation and caches the best result via MIOPEN_USER_DB_PATH.

First inference will be slower while benchmarking; subsequent calls use cache.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

engine.py

@@ -45,6 +45,15 @@ def load_model() -> bool:
             _is_turbo = False
 
         _sample_rate = 24000
+
+        # Enable MIOpen algorithm benchmarking. Without this, PyTorch picks
+        # convolution algorithms heuristically and passes ptr=0/size=0 workspace
+        # to MIOpen, forcing a slow fallback on every conv op. With benchmark=True,
+        # PyTorch evaluates algorithms with proper workspace on first use and caches
+        # the best result (persisted via MIOPEN_USER_DB_PATH volume mount).
+        if torch.cuda.is_available():
+            torch.backends.cudnn.benchmark = True
+
         _patch_timing(chatterbox_model)
         logger.info("Model loaded successfully")
         return True