Consolidate updates to dependencies

[gaby]

• Consolidates updates to several python and github-actions dependencies.
• Updated docs to use docker compose instead of docker-compose
• Added py3.12 and pypy-3.10 to CI workflow
• Formatted code base with black
• highest supported pydantic is 2.4.x. Anything higher than this fails to run with mypy errors.
• Updating mypy fails.

Closes #577 #588 #548 #567 #556 #554 #564 #563 #560 #495 #518

Tests fail with: #497
Related to #596

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 75.00000% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 77.89%. Comparing base (1213ca7) to head (ea59eea).

Files	Patch %	Lines
aredis_om/model/query_resolver.py	0.00%	2 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #597      +/-   ##
==========================================
+ Coverage   77.87%   77.89%   +0.01%     
==========================================
  Files          15       15              
  Lines        1279     1280       +1     
==========================================
+ Hits          996      997       +1     
  Misses        283      283              

Flag	Coverage Δ
unit	77.89% <75.00%> (+0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[slorello89]

There's something very odd going on in the sync versions for python 3.12, looks like it's improperly escaping whitespace & punctuation:

"FT.SEARCH" "redis-om:testing:0.05308182803634398:member:index" "@first_name:{'Andrew\\\\,\\\\ the\\\\ Michael'} " "LIMIT" "0" "1"

[slorello89]

Looks like the issue was caused by some updates to fstrings in python 3.12, to match the redis tag syntax, we naturally needed to escape braces in interpolated strings, see an example here, this is all well and good, and it still works in python 3.12, however we regenerate model.py using unasync, and the std_tokenizer changed in 3.12 so that f"@{field_name}:{{{value}}}" is yielded as a series of tokens @, {, field_name, }, :{, {, value, }, }, which from a code-interpretion perspective, is correct, however when the tokens are being reorganized they come out as `f"@{field_name}:{ {value}}", notice the space and missing braces.

It does not appear that there is a way to work around this with the code as-is, and changing those f-strings to the non-condensed format methods seems to resolve the tokenization issues:

so:

result += f"@{field_name}:{{{value}}}"

becomes:

result += "@{field_name}:{{{value}}}".format(
  field_name=field_name, value=value
)

This is a bit of hack, but it gets the job done and is the shortest and simplest way to the result we're looking for here, thoughts @banker, @gaby, @uglide?

[slorello89]

Approving, should get the OK from either @banker or @uglide regarding #597 (comment)

[slorello89]

Spoke with @banker we're good to go (when version update tests pass!)