forge_space_photo_maker_v2, bf16 is only supported on A100+ GPUs

`fa2F@v2.6.3-24-gbdf733b` is not supported because:
requires device with capability > (8, 0) but your GPU has capability (7, 5) (too old)
bf16 is only supported on A100+ GPUs
`cutlassF-pt` is not supported because:
bf16 is only supported on A100+ GPUs

我在运行 PhotoMaker V2 时，报错 GPU 不支持，然后我改了文件 `webui\extensions-builtin\forge_space_photo_maker_v2\forge_app.py` 第 40 行 :
```
if torch.cuda.is_available() and torch.cuda.is_bf16_supported():
torch_dtype = torch.bfloat16
else:
torch_dtype = torch.float16
```
改为
```
if torch.cuda.is_available() and torch.cuda.is_bf16_supported():
torch_dtype = torch.float16
else:
torch_dtype = torch.float16
```
就可以了，我不知道哪里的问题，竟然蒙对了，期望敏神修复它，提一个 Issue，希望对遇到相同报错的朋友有用

Start in webui\extensions-builtin\forge_space_photo_maker_v2\forge_app.py around line 40 and inspect how the torch dtype is selected. Run PhotoMaker V2 on a GPU with capability 7.5 to reproduce the reported bf16 error, then verify that the selected dtype avoids the unsupported A100-only path and the workflow completes.

Written by the indexing model from the issue text.