Fix issue with assertion caused by pluggable schema validator

[dmontagu]

Fix issue with type adapters present when plugins are installed.

[cloudflare-workers-and-pages]

Deploying pydantic-docs with    Cloudflare Pages

Latest commit:	47e20ac
Status:	✅  Deploy successful!
Preview URL:	https://e5e4e2e4.pydantic-docs.pages.dev
Branch Preview URL:	https://dmontagu-fix-pluggable-schem.pydantic-docs.pages.dev

View logs

[dmontagu]

This is where the problem is — because we were asserting that it was a SchemaValidator, but it could be a PluggableSchemaValidator. While we could just drop this assert, @adriangb wanted to just stop lying in the create_schema_validator return type annotation.

[dmontagu]

(And I agree with that choice, given I wasn't sure where else there might be problems.)

Ideally we would make it so that a PluggableSchemaValidator was actually a subclass of SchemaValidator, or similar, but anyway this works for now.

[dmontagu]

This raises an error on main, so obviously it's a problem lol

[codspeed-hq]

CodSpeed Performance Report

Merging #9838 will not alter performance

_{Comparing dmontagu/fix-pluggable-schema-validator-assertion (47e20ac) with main (cd3b0ad)}

Summary

✅ 13 untouched benchmarks

[dmontagu]

One other super annoying thing that we should fix soon is that you can't run the plugins tests on their own. I'm not sure where we are affecting state that causes them to pass when you run pytest tests but to fail when you run pytest tests/test_plugins.py but ... I think that should get fixed.

I'll note that if you change the _plugins: ... = None to _plugins: ... = {} this goes away and all tests pass. BUT I think we have special logic when _plugins is None, so maybe we want to add a test that actually ensures this behavior is respected?

[sydney-runkle]

Great, thanks for this quick fix!

[github-actions]

Coverage report

This PR does not seem to contain any modification to coverable code.