Disconnect the connection during the video generation process using wan2.2

### 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

I used ComfyUI_windows_portable_nvidia.7z downloaded on August 2, 2025. I loaded the built-in video_wan2_2_14B_t2v and video_wan2_2_14B_i2v workflows of comfyui and ran them. My custom_nodes only have comfyui-manager and comfyui-kjnodes. I hope to be able to generate videos from text and images normally

### Actual Behavior

### Steps to Reproduce

Download the latest ComfyUI_windows_portable_nvidia.7z with wan2.2 workflow, load the built-in video_wan2_2_14B_t2v and video_wan2_2_14B_i2v workflows of comfyui and run them. No custom nodes are required。The problem occurs after the sampling of the high-noise process is completed or when the wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors model is loaded in the low-noise process

### Debug Logs

```powershell
D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --fast fp16_accumulation
[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2025-08-02 08:13:15.356
** Platform: Windows
** Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
** Python executable: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\python_embeded\python.exe
** ComfyUI Path: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI
** ComfyUI Base Folder Path: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI
** User directory: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\user
** ComfyUI-Manager config path: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini
** Log path: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\user\comfyui.log

Prestartup times for custom nodes:
2.3 seconds: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager

Checkpoint files will always be loaded safely.
Total VRAM 16303 MB, total RAM 32104 MB
pytorch version: 2.7.1+cu128
Enabled fp16 accumulation.
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5070 Ti : cudaMallocAsync
Using pytorch attention
Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
ComfyUI version: 0.3.47
ComfyUI frontend version: 1.23.4
[Prompt Server] web root: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static
### Loading: ComfyUI-Manager (V3.35)
[ComfyUI-Manager] network_mode: public
### ComfyUI Revision: 150 [2f74e179] *DETACHED | Released on '2025-07-29'

Import times for custom nodes:
0.0 seconds: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
0.1 seconds: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-kjnodes
0.6 seconds: D:\ComfyUI_windows_portable_nvidia_0802\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager

[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
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
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
FETCH ComfyRegistry Data: 5/93
got prompt
Using pytorch attention in VAE
Using pytorch attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
Using scaled fp8: fp8 matrix mult: False, scale input: False
Requested to load WanTEModel
loaded completely 9.5367431640625e+25 6419.477203369141 True
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cuda:0, dtype: torch.float16
FETCH ComfyRegistry Data: 10/93
FETCH ComfyRegistry Data: 15/93
Requested to load WanVAE
loaded completely 269.8747215270996 242.02829551696777 True
FETCH ComfyRegistry Data: 20/93
Using scaled fp8: fp8 matrix mult: True, scale input: True
model weight dtype torch.float8_e4m3fn, manual cast: torch.float16
model_type FLOW
FETCH ComfyRegistry Data: 25/93
FETCH ComfyRegistry Data: 30/93
Requested to load WAN21
FETCH ComfyRegistry Data: 35/93
loaded partially 5971.982173297118 5971.978355407715 0
0%| | 0/3 [00:00pause
请按任意键继续. . .
```

### Other

_No response_