docs: update badges

[casperdcl]

• add |Choco|
• add |Brew|
• fix style for |Conda|
• add |PyPI|
• add |Packages|
• add a Chocolatey section
• reorder install options sections based on
  • number of users
  • number of potential users <-- I'd recommend this
  • number of desired users
• fixes ref: import-url/update --jobs affects uploading with --to-remote dvc.org#3233
• related install: reorder methods and add badges #7495

[skshetry]

Do we really need this number of badges? Maybe, we could just put a same page link to Installation section?

[efiop]

@skshetry Good point. So far the idea was that it is conveinient to see all of those badges, as you are intuitively looking for them whenever you are checking out a new project. Hiding them will make it less visible, so I'm not sure if it is a good thing. It does get a bit crowded, you are right.

[casperdcl]

Indeed, though if we want to drop things arguably |Release| |CI| |Maintainability| |Coverage| are least relevant to end users.

[efiop]

Ok, looking at readme now, I do see that it is way too crowded with installation badges :(

Indeed, though if we want to drop things arguably |Release| |CI| |Maintainability| |Coverage| are least relevant to end users.

That is not true, maintainability/coverage/ci all matters, as it is a sign of a well-maintained project. Those should not be removed.

[casperdcl]

Doesn't look too crowded to me:

Showing we support different installation options means more users likely to bother reading on an actually install. Maintainability/coverage/ci (in terms of users, not devs) are either completely ignored or assumed to be perfect - in both cases they don't necessarily need to be at the top of the readme if we want to drop some badges.

[skshetry]

I feel that 19 links at the top is too many. It makes it hard to scan through when the objective of having top-links and badges is to be quick and easy.