Design addition and updates to archived data

[justincc]

See the parent epic for the requirements. The immediate need is to add to or update the Meyer dataset but we should do this with an eye to reusing the same code for future additions and updates.

The step here is to come up with an initial design for doing these additions and updates through the ingest system. This will involve talking to @mshadbolt and @clairerye to clarify requirements and then to come up with a design for how we could implement them in our system.

Acceptance Criteria

• [ ] There is a proposed design that rests on wrangler requirements.
• [ ] This design has been reviewed by the team and feedback incorporated.

[aaclan-ebi]

Initial design doc: https://docs.google.com/document/d/1XDbEnct82WZOzzfPoOCPjPWuwYKJyLyqm4H5A1N88JU/edit#

This is still in progress but feel free to have early comments.

[aaclan-ebi]

This is heavily dependent on the archiving flow. We need to finalise that first in order to proceed with the design of the updates.