fix(gguf): ensure dequantized tensors are on correct device for MPS

[Pfannkuchensack]

Summary

When using GGUF-quantized models on MPS (Apple Silicon), the dequantized tensors could end up on a different device than the other operands in math operations, causing "Expected all tensors to be on the same device" errors.

This fix ensures that after dequantization, tensors are moved to the same device as the other tensors in the operation.

Related Issues / Discussions

(https://discord.com/channels/1020123559063990373/1149506274971631688/1454480237311168654)

QA Instructions

Test with z_image_turbo-Q4_K.gguf, Qwen_3_4b-Q6_K.gguf on mac

Merge Plan

No big change.

Checklist

• The PR has a short but descriptive title, suitable for a changelog
• Tests added / updated (if applicable)
• ❗Changes to a redux slice have a corresponding migration
• Documentation added / updated (if applicable)
• Updated What's New copy (if doing a release after this PR)

[Vargol]

Note this issue doesn't occur with keep_ram_copy_of_weights enabled as it defaults on enabled you'll need

keep_ram_copy_of_weights: False

in the invoke.yaml when testing
Also I believe partial loading can work around the issue too, either of those settings being enabled are sub-optimal on MPS.

It's basically the same issue as #7939

Oh, I believe it breaks on CUDA with the same settings, If you've got the VRAM to run it.

[gogurtenjoyer]

I've tested this PR on a M5 with 32gb RAM. It allows generations, with no errors, with keep_ram_copy_of_weights: False set. No errors reported in console, and the speed is the same as before.

[lstein]

Looks good. I'm going ahead with an approval to merge.