fix: sanitize target_flat — clips are inference tensors from outer inference_mode

The clips list is built inside ComfyUI's inference_mode context, so every
element is an inference tensor. torch.stack().clone() propagates the flag.
Use zeros+copy_ (same pattern as params/buffers) to get a normal tensor,
so mel_converter(target_flat) inside no_grad produces a saveable input.

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

nodes/selva_bigvgan_trainer.py

@@ -268,8 +268,17 @@ class SelvaBigvganTrainer:
                             start = random.randint(0, clip.shape[0] - segment_samples)
                             batch.append(clip[start : start + segment_samples])
 
-                        target_flat = torch.stack(batch).to(device, dtype).clone()  # [B, T]
-                        target_wav  = target_flat.unsqueeze(1)                # [B, 1, T]
+                        # clips were loaded in ComfyUI's outer inference_mode, so every
+                        # element is an inference tensor. torch.stack().clone() is still
+                        # an inference tensor (the flag propagates through all ops).
+                        # Use zeros+copy_ to produce a genuine normal tensor.
+                        _stacked = torch.stack(batch).to(device, dtype)
+                        target_flat = torch.zeros(
+                            _stacked.shape, device=device, dtype=dtype
+                        )
+                        target_flat.copy_(_stacked)
+                        del _stacked
+                        target_wav = target_flat.unsqueeze(1)                 # [B, 1, T]
 
                         # Fixed target mel — buffers are now normal tensors (sanitized
                         # above), so torch.no_grad() correctly produces a non-inference,