[PLT-1725] Changed ontology dataclasses to pydantic dataclasses

[Gabefire]

Description

• Pydantic V2 has a standin replacement for python dataclasses but with more features and better validation.
• Changed ontology creation to take advantage of pydantic features

For example for this code:

classification_features = [
    lb.Classification(
        class_type="test",
        name=1,
        options=[
            lb.Option(value="blurry", label="Blurry"),
            lb.Option(value="distorted", label="Distorted"),
        ],
    )
]

ontology_builder = lb.OntologyBuilder(classifications=classification_features).asdict()

class_type and name is given the wrong type for the classification feature (string instead of an enum value and int instead of string). Before change, it would give this error

With pydantic validation it gives this error (catching the name being an int as well)

Type of change

Please delete options that are not relevant.

• New feature (non-breaking change which adds functionality)

All Submissions

• Have you followed the guidelines in our Contributing document?
• Have you provided a description?
• Are your changes properly formatted?