Fix print_frames_info logging garbage bound-method reprs for tensor frames

torch.Tensor has .size()/.mode() methods with the same names as PIL
Image's .size/.mode attributes, which print_frames_info reads via
getattr assuming a PIL Image. Since #53 the RIFE->upscale tensor
handoff can pass a tensor frame list to this same logging call,
producing e.g. "pil_size=<built-in method size of Tensor object at
0x...>" in production logs instead of a clean None.

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

logging_utils.py

@@ -71,8 +71,13 @@ def print_frames_info(label: str, frames: Any, fps: Any) -> None:
     kind = type(first).__name__ if first is not None else "n/a"
     shape = getattr(first, "shape", None)
     dtype = getattr(first, "dtype", None)
-    size = getattr(first, "size", None)  # PIL Image
-    mode = getattr(first, "mode", None)  # PIL Image
+    # PIL Image exposes .size/.mode as plain attributes; torch.Tensor happens to have same-named
+    # *methods* (Tensor.size(), Tensor.mode()) — skip those so callers passing tensors (e.g. the
+    # RIFE->upscale GPU handoff) don't log a garbage bound-method repr in their place.
+    size = getattr(first, "size", None)
+    size = size if not callable(size) else None
+    mode = getattr(first, "mode", None)
+    mode = mode if not callable(mode) else None
     print(
         f"[export] {label}: n={n}, fps={fps}, frame_type={kind}, shape={shape}, dtype={dtype}, "
         f"pil_size={size}, pil_mode={mode}",