fix: URL-encode document names in document index client requests

[Michael-JB]

Description

Prior to this commit, we did not URL-encode document names when querying
the document index. This caused problems when names have special
characters, i.e., '/' or '?'. We do not need to URL-encode other parts
of the URL (e.g., namespace or collection) as these are already
constrained to URL-safe characters. In addition to this fix, this commit
includes the following changes:

• update DocumentPath.from_slash_separated_str to correctly delimit
  paths whose document name contains a forward slash. Ideally we'd
  remove both this and DocumentPath.to_slash_separated_str in the
  future.
• parametrize tests to include special characters and increase coverage
• migrate to new document index collection to avoid CI conflicts

Before Merging

• Review the code changes
  • Unused print / comments / TODOs
  • Missing docstrings for functions that should have them
  • Consistent variable names
  • ...
• Update changelog.md if necessary
• Commit messages should contain a semantic label and the ticket number
  • Consider squashing if this is not the case

[NiklasKoehneckeAA]

If we want to remove the DocumentPath.from_slash_separated_str, we could raise a deprecated warning in this

[Michael-JB]

If we want to remove the DocumentPath.from_slash_separated_str, we could raise a deprecated warning in this

I only suggest removing them because a slash is a valid character in a document name, which makes unescaped slashes a bad choice for delimiting.

I'm not sure what the use-case for from_slash_separated_str is, and I imagineto_slash_separated_stris only really useful for logging purposes. We don't use them internally either, hence my suggestion to remove them. What do you think?

Edit: in either case, I think it would make sense to do this in a separate PR.