Regional Lora Hooks don't work on Flux/Quantized flux versions

[Matheus-Torres-Tavares]

Any chance of this being supported in the future? The regional Lora hooks are a game changer and one of the best features of this pack.
For more context:

Anytime I run GGUF fp8 flux I get the following errors:

ERROR lora diffusion_model.double_blocks.9.img_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.img_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.img_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.txt_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.txt_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.txt_attn.qkv.weight shape '[3072, 3264]' is invalid for input of size 9437184
ERROR lora diffusion_model.double_blocks.9.img_attn.proj.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.img_mod.lin.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.txt_mod.lin.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.txt_attn.proj.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.img_mlp.0.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.img_mlp.2.weight The size of tensor a (13056) must match the size of tensor b (12288) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.txt_mlp.0.weight The size of tensor a (3264) must match the size of tensor b (3072) at non-singleton dimension 1
ERROR lora diffusion_model.double_blocks.9.txt_mlp.2.weight The size of tensor a (13056) must match the size of tensor b (12288) at non-singleton dimension 1

If I run flux dev the workflow runs but doesn't apply any of the regional Loras.

[Kosinkadink]

This should be fixed when my hooks become a native feature in my ComfyUI PR, currently estimated to get merged around Thanksgiving: comfyanonymous/ComfyUI#5583

My corresponding ADE PR will make use of these hooks in the existing nodes, and also deprecate them in favor of the new native nodes. #498