DM-42302: Drop support for Pydantic v1.

[TallJimbo]

Requires lsst/daf_butler#929

Checklist

• ran Jenkins
• added a release note for user-visible changes to doc/changes

[codecov]

Codecov Report

Attention: 32 lines in your changes are missing coverage. Please review.

Comparison is base (9d3cecc) 82.51% compared to head (473a15e) 82.25%.

Files	Patch %	Lines
...ython/lsst/pipe/base/tests/mocks/_storage_class.py	15.38%	20 Missing and 2 partials ⚠️
python/lsst/pipe/base/_task_metadata.py	23.07%	9 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #396      +/-   ##
==========================================
- Coverage   82.51%   82.25%   -0.26%     
==========================================
  Files          92       92              
  Lines       10430    10460      +30     
  Branches     1971     1977       +6     
==========================================
- Hits         8606     8604       -2     
- Misses       1483     1511      +28     
- Partials      341      345       +4     

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

[TallJimbo]

I have no idea why the doc build is trying to build docs for Pydantic itself (that seems to be what it's failing on), or why it's a problem here but not in daf_butler.

[TallJimbo]

I've added some trivial forwarders to public classes that inherit from pydantic.BaseModel, to just override the docstrings so they aren't inherited. It's not at all pretty, but it gets us through without having to weaken our guards against problems in our own docs.

It's hard for me to believe that automodapi is really designed to assume that it should parse the docstrings from any third-party classes that inherited from in its targeted package as rST, with no way to turn it off, but I can find no evidence that anyone else has run into this before.

[timj]

The sphinx fixes are pretty bad but seem pragmatic for now.