Change dataclasses.dataclass to pydantic.dataclasses.dataclass

[deanpoulos]

This change is motivated by the following use-case.

Let's say I create a QuAM object to represent my QPU. Since every QuAM component is also a @quam_dataclass, I can use Pydantic's dataclass scheme generation feature as follows:

Do this to generate a schema:

schema = TypeAdapter(QuAM).json_schema()
with open("schema.json", "w+") as f:
   json.dump(schema, f, indent=2))

Then, if you load the schema into my IDE using a JSON schema mapping, you get required-field checking, type-checking, attribute hints, etc.
trimmed_output.webm

I believe we can enhance the number of schema features we exploit by doing something like this:

@quam_dataclass
class XYZ:
        age: Optional[int] = pydantic.dataclasses.field(metadata=dict(
            description="some description"
        )
    )

but I haven't tested it.

Could you please run this against tests and consider whether it will break fundamental features of QuAM? If not, I believe it would be a great improvement.

[nulinspiratie]

If it doesn't break anything, this is probably a pretty major improvement for QUAM, great idea!
I'm a bit surprised that it works, and it also seems to be okay with references??
Very strange, I will investigate.