Add model based on DeepDanbooru

[MitPitt]

DeepDanbooru is a caption model trained on tags from boorus. Mostly anime, but it does work incredibly well with non-anime images too. Booru-style prompts (example: 1girl, simple background, watercolor, ...) work very well in text2img.

DeepDanbooru is originally used on danbooru to tag images, but now is also used to fine-tune SD.
Since then a couple modifications (like removing underscores in tags) have been made to it, to work better with SD.

See demo and links to models here:
https://huggingface.co/spaces/SmilingWolf/wd-v1-4-tags

[theovercomer8]

Will definitely consider doing soon.