Migrate reuseable content from docs to package docs / website

[dlebauer]

Consider breaking this into sub-tasks. See also, comment below.

• Technical docs
  • Tables - define tables and fields (only those that are included in this package)
  • Schema diagram should be published to (and replace broken) betydb.org/schemas
  • Constraints (Implement (migrate) constraints and validation from postgres betydb #14)
  • basically nothing else there is a lot of outdated ruby/postgres/technical docs
• data access ... mostly this is already in vignettes, but there is likely additional content to migrate
• data entry
  • procedure related to extracting and entering data from papers ... this should use an updated version of the betydb template spreadsheet
  • when there is a use case implement basic new data ingest functionality Tools to import data #18
  • do not re-implement all possible ways to upload data (UI, API)
  • quality control; error reporting

Out of scope:

• any new functionality, including
  • re-creating UI, API etc
  • new data upload functionality (can be separate issue, and should focus on spreadsheet --> database)
  • these should be separate issues, and at present are very low priority
  • interactive schema

[dlebauer]

Follow-up from review of #12. Using this issue as the umbrella for structure and documentation work that grows out of that review.

• Publish a schema / join map for packaged tables.
• Document authoritative tables and join keys for common workflows.
• Reuse canonical metadata to keep package docs and website content in sync.
• Clarify how packaged tables differ from source CSV metadata, if both are kept.
• Add user-facing guidance for interpreting IDs and relationship tables. (can be wrapped into Implement machine-readable source of truth for data constraints and model validation rules #22)

PR-specific blockers from #12 should stay on the PR review rather than move here.