Feature Request: Separate boomark types

[rocketpear]

Afaik, currently, each bookmark is of one "type." One cool feature mymind has is that it treats different kinds of content differently. If you save an article, it extracts the content and when you open that bookmark, it displays the article text in reading mode by default. If you just save a website, it displays the screenshot of the website by default. If you save a recipe, it extracts just ingredients and steps and displays that when you open the bookmark. If you save a photo or a screenshot, it adds the color palette by default and runs OCR. You get the idea.
It also indicates the type of content on the bookmark card itself (showing if it's a reddit post, a tweet, a video etc).

I feel like sepatrating these bookmark types might be helpful because it allows for several things:

1. Treating different types of content differently when scraping, leading to more accuracy (eg. knowing exactly the fields users are interested in when saving a tweet as opposed to newspaper article).
2. Ability to surface the most desired functionality based on content (reading mode vs screenshot vs color palette etc).
3. Ability for users to tweak behaviour on per-type basis. Like if there is option to choose between making full site backup or a screenshot automatically in the future, users could set this differently for just "website links" and news articles.