GGUF Q8_0 and Automatic Diffusion (8bit LoRa) generating poor quality

Nobody has claimed this yet.

Since ba01ad37, LoRas loaded in 8bit to the Q8_0 GGUF generate to a poor quality. Loading the LoRa in 16bit appears to fix this issue, but there are subtle differences in the generations from rounding.

This does not seem to happen with FP8 safetensor or with the NF4 - just the Q8_0 GGUF.
This does not happen with checkpoint ba01ad37 and earlier.

Example at commit 3b9b2f65:
` A pixelart drawing of a chicken`

Diffusion in Low bits set to Automatic:

Diffusion in Low bits set to Automatic (FP16 Lora)

Example at commit ba01ad37 (last testable commit before 8bit LoRa changed):

Diffusion in Low bits set to Automatic:

Diffusion in Low bits set to Automatic (FP16 Lora):

Generations with the FP8 safetensor for comparison:

Diffusion in Low bits set to Automatic:

Diffusion in Low bits set to Automatic (FP16 Lora):

And here is the NF4 for comparison:

Diffusion in Low bits set to Automatic:

Diffusion in Low bits set to Automatic (FP16 Lora):

(LoRa used here was https://civitai.com/models/704779/flux-gen-5-trainer-sprites but appears to happen with all LoRas)

Reproduce the example with Q8_0 GGUF, Automatic diffusion low bits, and an 8bit LoRa at commits 3b9b2f65 and ba01ad37, then compare the generated results with FP16 LoRa, FP8 safetensor, and NF4. Done means the Q8_0 GGUF 8bit LoRa generation no longer has the reported quality degradation without requiring the FP16 workaround.

Written by the indexing model from the issue text.