Failure at the same spot after resuming the download

## Describe the bug

While attempting to download the Juggernaut-XL-v9 model, the process failed at around 27%. I restarted the download, and it resumed from 1.5 GB (roughly 24%) but failed again at 27%. A third attempt led to the same frustrating result: resumption from 1.5 GB and failure at 27% once more.

Hoping to resolve the issue, I closed the software, completely deleted the folder: "C:\Program Files\AI Playground\resources\service\models\stable_diffusion\checkpoints\RunDiffusion---Juggernaut-XL-v9_tmp", and tried downloading the model again.

This time, the download did perform "better", failing at about 42%. When I restarted the process, it resumed from 1.3 GB, even less than last time. And just as I feared, it failed at 42% once again.

To confirm my suspicion, I restarted the download once more, and sure enough, it still resumed from 1.3 GB. Rather than waiting for another inevitable failure at 42% like a fool, I decided to report the issue here.

## To Reproduce

Steps to reproduce the behavior:
1. Start downloading Juggernaut-XL-v9.
2. Wait until approximately 27% is completed, then cancel the download (note: this may behave differently from an automatic failure).
3. Restart the download for the same model—it will probably fail at the same point where you previously canceled.
4. Try again, and the download will resume from the same point as before. If you wait, it will fail at the exact same spot once more.

## Environment (please complete the following information):

- OS: [Windows11]
- GPU: [Intel Arc Graphics 128 MB]
- CPU: [Ultra 5 125H]
- Version: [v2.0.4b-MTL-H]