Disable warnings from model_dump

[alicederyn]

Pull Request Checklist

• Fixes #
• Tests added
• Documentation/examples added
• Good commit messages and/or PR title

Description of PR
Currently, using Pydantic V2 with the Input/Output mixin types and a non-str field prints a UserWarning when the class is declared, of the form "UserWarning: Pydantic serializer warning: Expected int but got str with value '{{...}}' - serialized value may not be as expected". This is caused by using model_dump on an object constructed with all string values, i.e. one which does not pass the type-checking performed by the method, as part of class creation.

This PR just adds warnings="none" to the two uses of model_dump, to disable these warnings.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.0%. Comparing base (d553546) to head (4037c57).
Report is 154 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #1214     +/-   ##
=======================================
+ Coverage   81.7%   86.0%   +4.2%     
=======================================
  Files         54      60      +6     
  Lines       4208    4040    -168     
  Branches     889     840     -49     
=======================================
+ Hits        3439    3475     +36     
+ Misses       574     392    -182     
+ Partials     195     173     -22     

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

[alicederyn]

I'm in two minds about this fixture — it does catch this issue and prevent any similar issues in future, but I don't like having to suppress the one deprecation warning that Pydantic appears to have accidentally left as a UserWarning. LMK if you'd rather leave it out, it's a separate commit so easy to revert.

[elliotgunton]

I would rather not optimise early here as it might get difficult/annoying when we move Hera's core code to Pydantic V2

[alicederyn]

Sure thing. Removed the fixture.