Bluesky post extractor

[GiovanH]

Functional bluesky downloader.

Example:

py -3.11 -m gallery_dl --verbose "https://bsky.app/profile/im.giovanh.com/post/3kaxkwevkn626" --verbose
[gallery-dl][debug] Version 1.26.2-dev - Git HEAD: b029f25f
[gallery-dl][debug] Python 3.11.5 - Windows-10-10.0.19045-SP0
[gallery-dl][debug] requests 2.31.0 - urllib3 1.26.16
[gallery-dl][debug] Configuration Files []
[gallery-dl][debug] Starting DownloadJob for 'https://bsky.app/profile/im.giovanh.com/post/3kaxkwevkn626'
[bluesky][debug] Using _BlueskyPostExtractor for 'https://bsky.app/profile/im.giovanh.com/post/3kaxkwevkn626'
[urllib3.connectionpool][debug] Starting new HTTPS connection (1): cdn.bsky.app:443
[urllib3.connectionpool][debug] https://cdn.bsky.app:443 "GET /img/feed_fullsize/plain/did:plc:kjx6y3groxh3sy5tkfyji6sy/baf
kreif5rwbbx3nrz4v3kcaqjycjohie7lkaempzffqu4ggqehww34jjpe@jpeg HTTP/1.1" 200 203359
* .\gallery-dl\bluesky\im.giovanh.com\3kaxkwevkn626 bafkreif5rwbbx3nrz4v3kcaqjycjohie7lkaempzffqu4ggqehww34jjpe.jpg

TODO;

• Replace netrc with gallery-dl native authentication
• Remove chitose dependency (???? @mikf)
• Metadata postprocessor should store metadata by-post by default. Instead, it's targeting a global metadata.json file. @mikf this looks like there's pathfmt data missing based on what I see in MetadataPP; what's the best way to set a default for this?
• Add readme

[jhwgh1968]

Just a drive by comment:

In the time since this PR was created, they have added support for the viewing of profiles while not logged in.

Poking at this a tiny bit, it looks like this API call used by their web interface would allow downloading by "user page":

https://public.api.bsky.app/xrpc/app.bsky.feed.getAuthorFeed?actor=did:plc:[user-did]&filter=posts_and_author_threads&limit=30

Hopefully @mikf looks at the weird build failures soon! (I take it you've rebased recently, @GiovanH?)