update TableRecord type annotation to accept None values

[sryza]

Summary & Motivation

This reflects the runtime type-checking we're doing inside the constructor.

This was causing problems, because we were loading values from the DB that had None values and then an outer class that was a Pydantic model was using the type annotations to validate them, and erroring.

records.0.0.extension.bool
  Input should be a valid boolean [type=bool_type, input_value=None, input_type=NoneType]
    For further information visit https://errors.pydantic.dev/2.6/v/bool_type

How I Tested These Changes

[alangenfeld]

would be good to add a test that would have caught this

[alangenfeld]

outer class that was a Pydantic model was using the type annotations to validate them

can use InstanceOf at the pydantic -> NT barrier to prevent shenanigans https://github.com/dagster-io/dagster/pull/21702/files

[sryza]

would be good to add a test that would have caught this

I started to do this but it felt a little goofy. I.e. it would basically be:

• Construct one of these objects with a None value
• Stick it inside a TableMetadataValue

Which is not a a hard test to write, but I think I'd scratch my head a bit if I came across it - wondering why we don't have some version of this for every place we're checking types, which feels a little absurd.

Maybe I'm overthinking it.