Switch model inference from no_grad to inference_mode

Both call sites convert outputs to numpy immediately, so the stricter
inference-mode tensors are safe; saves autograd view/version tracking
overhead. Verified end-to-end on CUDA (alpha finite, no NaN).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

bgfilter/segmentation.py

@@ -62,7 +62,7 @@ class BiRefNetSegmenter:
         """Return a soft foreground mask, float32 0..1, at the input resolution."""
         image = Image.fromarray(rgb.astype(np.uint8), mode="RGB")
         tensor = self.transform(image).unsqueeze(0).to(self.device)
-        with self.torch.no_grad():
+        with self.torch.inference_mode():
             pred = self.model(tensor)[-1].sigmoid()
         pred = pred[0, 0].detach().float().cpu().numpy()
         if pred.shape != rgb.shape[:2]:

bgfilter/vitmatte_infer.py

@@ -36,7 +36,7 @@ class ViTMatteRunner:
         trimap_image = Image.fromarray(trimap.astype(np.uint8), mode="L")
         inputs = self.processor(images=image, trimaps=trimap_image, return_tensors="pt")
         inputs = {key: value.to(self.device) for key, value in inputs.items()}
-        with self.torch.no_grad():
+        with self.torch.inference_mode():
             outputs = self.model(**inputs)
 
         alpha = outputs.alphas[0, 0].detach().float().cpu().numpy()