Server does not start with pytorch nightly

### Custom Node Testing

- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

Expecting starting server after running python main.py or error messages.

I am using a 50 series graphic card.
Server does not starting when I was using pytorch nightly
`pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu129`
No warning or error messages

### Actual Behavior

I am using a 50 series graphic card.
Server does not starting when I was using pytorch nightly
`pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu129`
No warning or error messages

**Using Stable pytorch fixed the problem**
```powershell
pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu129
...
...
Successfully installed torch-2.8.0+cu129
(comfy) PS D:\git_repository\ComfyUI> python .\main.py
Checkpoint files will always be loaded safely.
Total VRAM 16303 MB, total RAM 65349 MB
pytorch version: 2.8.0+cu129
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5080 : cudaMallocAsync
Using pytorch attention
Python version: 3.12.11 | packaged by Anaconda, Inc. | (main, Jun 5 2025, 12:58:53) [MSC v.1929 64 bit (AMD64)]
ComfyUI version: 0.3.51
****** User settings have been changed to be stored on the server instead of browser storage. ******
****** For multi-user setups add the --multi-user CLI argument to enable multiple user profiles. ******
ComfyUI frontend version: 1.25.9
[Prompt Server] web root: D:\anaconda3\envs\comfy\Lib\site-packages\comfyui_frontend_package\static

Import times for custom nodes:
0.0 seconds: D:\git_repository\ComfyUI\custom_nodes\websocket_image_save.py

Context impl SQLiteImpl.
Will assume non-transactional DDL.
No target revision found.
Starting server

To see the GUI go to: http://127.0.0.1:8188
```

### Steps to Reproduce

```
(comfy) PS D:\git_repository\ComfyUI> python .\main.py
Checkpoint files will always be loaded safely.
Total VRAM 16303 MB, total RAM 65349 MB
pytorch version: 2.9.0.dev20250821+cu128
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5080 : cudaMallocAsync
(comfy) PS D:\git_repository\ComfyUI> python .\main.py
Checkpoint files will always be loaded safely.
Total VRAM 16303 MB, total RAM 65349 MB
pytorch version: 2.9.0.dev20250821+cu128
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5080 : cudaMallocAsync
```

### Debug Logs

```powershell
Successfully installed torch-2.8.0+cu129
(comfy) PS D:\git_repository\ComfyUI> python .\main.py
Checkpoint files will always be loaded safely.
Total VRAM 16303 MB, total RAM 65349 MB
pytorch version: 2.8.0+cu129
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5080 : cudaMallocAsync
Using pytorch attention
Python version: 3.12.11 | packaged by Anaconda, Inc. | (main, Jun 5 2025, 12:58:53) [MSC v.1929 64 bit (AMD64)]
ComfyUI version: 0.3.51
****** User settings have been changed to be stored on the server instead of browser storage. ******
****** For multi-user setups add the --multi-user CLI argument to enable multiple user profiles. ******
ComfyUI frontend version: 1.25.9
[Prompt Server] web root: D:\anaconda3\envs\comfy\Lib\site-packages\comfyui_frontend_package\static

Import times for custom nodes:
0.0 seconds: D:\git_repository\ComfyUI\custom_nodes\websocket_image_save.py

Context impl SQLiteImpl.
Will assume non-transactional DDL.
No target revision found.
Starting server

To see the GUI go to: http://127.0.0.1:8188
```

### Other

_No response_