Fix: Wrap LoRA loading and offloading in defer_i2v_loading check

When defer_i2v_loading=True (I2V mode without provided image), the code
sets pipe=None but then tried to call pipe.load_lora_weights() and
pipe.enable_model_cpu_offload() on None, causing AttributeError.

This fix wraps the LoRA loading and offloading configuration blocks
inside an 'if not defer_i2v_loading:' condition so they are skipped
when the I2V model loading is deferred until after image generation.

videogen

@@ -3938,6 +3938,8 @@ def main(args):
     
     timing.end_step()  # model_loading
     
+    # Only apply LoRA and offloading if we actually loaded the model (not deferred)
+    if not defer_i2v_loading:
         # Apply LoRA if this is a LoRA model
         if is_lora and lora_id:
             timing.begin_step("lora_loading")