Use a dedicated URL-handling library

[jwodder]

There are various places in the code that would benefit from the use of a modern URL-handling library, such as furl, purl, or yarl. These include:

• is_same_url() in delete.py
• extract_species() in metadata/util.py (yarl doesn't help here; see Clean up URL parsing in extract_species() #1416 instead)
• get_content_url() in dandiapi.py
• get_instance() in utils.py
• is_page2_url() in utils.py
• _get_instance() in utils.py
• joinurl() in utils.py (yarl doesn't help here)
• Most of dandi/dandiarchive.py? (I think rewriting this would be more trouble than it's worth)
• implementing the URL validation needed for publications in asset metadata #1308

We should use such a library to improve the code.

[yarikoptic]

good idea. Only yarl seems to be alive, the other two had no changes in years. And dang -- yarl already has precommit setup for use of codespell ;-)

[yarikoptic]

please proceed with making a decision on this one @jwodder to unblock

• publications in asset metadata #1308

[github-actions]

🚀 Issue was released in 0.61.0 🚀