use **kwargs and change svd() calling convention to make svd() reusable

[wkpark]

use **kwargs and change svd() calling convention to make svd() reusable

• add required attributes to model_org, model_tuned, save_to
• set "*_alpha" using str(float(foo))
• add --min_diff, --clamp_quantile args based on v21.8.6 bmaltais/kohya_ss#1332 bmaltais/kohya_ss@a9ec90c

(venv) F:\webui\webui\sd-scripts>python networks\extract_lora_from_models.py --model_org ..\stable-diffusion-webui\models\Stable-diffusion\v1-5-pruned-emaonly.safetensors --model_tuned ..\stable-diffusion-webui\models\Stable-diffusion\py0fp16.safetensors --save_to xx.safetensors --device cuda --min_diff 0.0001
loading original SD model : ..\stable-diffusion-webui\models\Stable-diffusion\v1-5-pruned-emaonly.safetensors
UNet2DConditionModel: 64, 8, 768, False, False
loading u-net: <All keys matched successfully>
loading vae: <All keys matched successfully>
loading text encoder: <All keys matched successfully>
loading tuned SD model : ..\stable-diffusion-webui\models\Stable-diffusion\py0fp16.safetensors
UNet2DConditionModel: 64, 8, 768, False, False
loading u-net: <All keys matched successfully>
loading vae: <All keys matched successfully>
loading text encoder: <All keys matched successfully>
create LoRA network. base dim (rank): 4, alpha: 4
neuron dropout: p=None, rank dropout: p=None, module dropout: p=None
create LoRA for Text Encoder:
create LoRA for Text Encoder: 72 modules.
create LoRA for U-Net: 192 modules.
create LoRA network. base dim (rank): 4, alpha: 4
neuron dropout: p=None, rank dropout: p=None, module dropout: p=None
create LoRA for Text Encoder:
create LoRA for Text Encoder: 72 modules.
create LoRA for U-Net: 192 modules.
Text encoder is different. 0.0001220703125 > 0.0001
calculating by svd
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 264/264 [00:28<00:00,  9.12it/s]
create LoRA network from weights
create LoRA for Text Encoder:
create LoRA for Text Encoder: 72 modules.
create LoRA for U-Net: 192 modules.
enable LoRA for text encoder
enable LoRA for U-Net
Loading extracted LoRA weights: <All keys matched successfully>
LoRA weights are saved to: xx.safetensors

[kohya-ss]

Thank you for this! Sorry for the delay!