A style adjust of Animagine XL V3

A style merge base on https://civitai.com/models/260267/animagine-xl-v3

This is an extremely, extremely, extremely insignificant adjustment of the Animagine XL V3 .

All rights, including license agreements, remain with Cagliostrolab.

1.reduced the black outline.

2.adjust color

3.some style if I like it

I want to release a Lora , but the training using Lycrois with preset=full , and I made some train&merge.

Instead of release a more than 4G "Lora" and very long readme ,it's better to just make the all train%mbw&lbw and release it.

Thanks for the Cagliostrolab fantisy work.