Move Data Product Registration code to FastAPI framework

[PriyaBasker23]

User Story In the role of a developer, I aim to establish an API endpoint for the registration of a data product.

Value / Purpose This initiative is crucial for facilitating connectivity with the frontend and ensuring the successful registration of data products.

End point /data-products method - Post Data product information will be in the body of the call

Proposal

• [ ] Extract the code from the Python base (refer to the initial FastAPI commits in the repository).
• [ ] Repurpose the code for integration into the existing lambda function.
• [ ] Develop a Pydantic model, aiming for a single model for each endpoint to eliminate optional fields.
• [ ] Incorporate unit tests, replicating them from the current application.
• [ ] Illustrate an integration example with the firebreak, showcasing the functionality of the endpoint.
• [ ] For the present, omit the inclusion of authentication requirements.
• [ ] Ensure Documentation of the endpoint is working
• [ ] Add approriate logging which will be useful debugging

Definition of Done

• [ ] Deploy and validate the functionality of the endpoint.
• [ ] Implement unit tests to ensure code integrity.
• [ ] Add integration tests to verify seamless operation.
• [ ] Confirm that registered data products can successfully transmit information to the designated S3 bucket.

[MatMoore]

@seanprivett I think we can probably close this for now. We've since changed direction from the original proposal:

• Decision to use Datahub as a backend instead of s3
• Decision to delay introducing the data product concept until after launching the catalogue

And we've also already explored this from the perspective of a concierge MVP

• https://github.com/ministryofjustice/data-catalogue-metadata/ - the current way to manage metadata ourselves
• https://github.com/ministryofjustice/data-platform/pull/3884 - spike for bringing in pydantic

During beta we may still need a similar ticket to this, if we develop custom UIs for registering/editing specific metadata items rather than using an automated ingestion source. But I don't think we are ready to plan that just yet.