Apply style fixes

Author: github-actions[bot]

src/diffusers/loaders/lora_pipeline.py

@@ -1976,7 +1976,9 @@ def _maybe_expand_transformer_param_shape_or_error_(
             if isinstance(module, torch.nn.Linear):
                 is_bnb_4bit_quantized = module.weight.__class__.__name__ == "Params4bit"
                 if is_bnb_4bit_quantized and not is_bitsandbytes_available():
-                    raise ValueError("The checkpoint seems to have been quantized with `bitsandbytes` (4bits). Install `bitsandbytes` to load quantized checkpoints.")
+                    raise ValueError(
+                        "The checkpoint seems to have been quantized with `bitsandbytes` (4bits). Install `bitsandbytes` to load quantized checkpoints."
+                    )
                 elif is_bnb_4bit_quantized:
                     module_weight = dequantize_bnb_weight(module.weight, state=module.weight.quant_state).data
                 else: