Update the pydantic models for datasets and files.

[Essmaw]

Apply suggested modifications from @pierrepo.

[pierrepo]

@Essmaw Nice. Let me know when I can review and merge.

[pierrepo]

In models/datasets.py, do you think dataset_project and dataset_id_in_project could be computed fields, automatically derived from dataset_repository and dateset_id_in_repository, but with the ability to manually provide dataset_project and dataset_id_in_project if needed?

For instance (inspired from documentation):

class DatasetMetadata(BaseModel):
    dataset_repository: DatasetRepository = Field(
        ...,
        description=(
            "Name of the source repository. "
            "Allowed values: ZENODO, FIGSHARE, OSF, NOMAD, ATLAS, GPCRMD."
        ),
    )

    dataset_id_in_repository: str = Field(
        ...,
        description="Unique identifier of the dataset in the source repository.",
    )

    @computed_field
    def dataset_project(self) -> DatasetRepository:
        """Name of the project."""
        return self.dataset_repository

    @dataset_project.setter
    def dataset_project(self, new_name: DatasetProject) -> None:
        self.dataset_project = new_name

    @computed_field
    def dataset_id_in_project(self) -> str:
        """ID of the project."""
        return self.dataset_id_in_repository

    @dataset_project.setter
    def dataset_id_in_project(self, new_id: str) -> None:
        self.dataset_id_in_project = new_id

[pierrepo]

This sounds good @Essmaw 🎉
Ready to merge?