Fix is_using_oneflow_backend check (#1112)

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

- **New Features**
- Enhanced backend detection logic for improved compatibility with the
OneFlow library.
- Added a function to check for OneFlow library availability and CUDA
support.

- **Bug Fixes**
- Improved messaging for cases when the OneFlow backend is not detected.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

onediff_comfy_nodes/modules/oneflow/utils/booster_utils.py

@@ -1,5 +1,7 @@
 from typing import Union
 
+import oneflow
+
 import torch
 from comfy import model_management
 from comfy.model_base import BaseModel, SVD_img2vid
@@ -9,6 +11,7 @@
     OneflowDeployableModule as DeployableModule,
 )
 from onediff.utils import set_boolean_env_var
+from onediff.utils.import_utils import is_oneflow_available
 
 from ..patch_management import create_patch_executor, PatchType
 
@@ -63,6 +66,15 @@ def set_environment_for_svd_img2vid(model: ModelPatcher):
 
 
 def is_using_oneflow_backend(module):
+    # First, check if oneflow is available and CUDA is enabled
+    if is_oneflow_available() and not oneflow.cuda.is_available():
+        print("OneFlow CUDA support is not available")
+        return False
+
+    # Check if the module
+    if isinstance(module, oneflow.nn.Module):
+        return True
+
     dc_patch_executor = create_patch_executor(PatchType.DCUNetExecutorPatch)
     if isinstance(module, ModelPatcher):
         deep_cache_module = dc_patch_executor.get_patch(module)
@@ -85,7 +97,17 @@ def is_using_oneflow_backend(module):
     if isinstance(module, DeployableModule):
         return True
 
-    raise RuntimeError("")
+    if hasattr(module, "parameters"):
+        for param in module.parameters():
+            if isinstance(param, oneflow.Tensor):
+                return True
+
+    warn_msg = (
+        f"OneFlow backend is not detected for the module, the module is {type(module)}"
+    )
+    print(warn_msg)
+    # If none of the above conditions are met, it's not using OneFlow backend
+    return False
 
 
 def clear_deployable_module_cache_and_unbind(