feat: support resolving did:peer:1 received in did exchange

[dbluhm]

This is an experimental implementation of supporting did:peer:1 resolution. This enables ACA-Py to correctly handle DID Exchange requests from AFJ (AFJ now fails to process our responses since we're not passing back a did:peer, still the unqualified values).

Notably, this does not support creating or sending did:peer:1, just storing and resolving.

Detailed changes:

• Storing a DID Document (using base connection manager) no longer first requires parsing as the (soon to be deprecated) DIDDoc class (not PyDID DIDDocument).
  • This alone removes a whole slue of issues that the LegacyCorrections class was written to deal with.
• Storing the key -> did -> connection mapping is now always delegated to BaseConnectionManager.record_did; this helps with consistently handling DIDs
• Fetching a DID Document with BaseConnectionManager.fetch_did_document no longer parses the response from the wallet using DIDDoc. This helps this call to be a primitive used by DID resolvers that rely on stored documents to resolve DIDs without the DIDDoc parsing logic getting in the way.
• Update the connection record to not require "Indy" shaped DIDs

Note to self, this resolver should have a clean up routine run when the associated connection record is deleted, similar to did:peer:2/3

[dbluhm]

This is now ready for review! I would like to continue building did:peer:2 support on top of these changes. @swcurran @usingtechnology

[swcurran]

Adding @Jsyro to reviewer. Can’t give a Maintainer approval, but I can translate… :-)

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

[usingtechnology]

LGTM, great explanation of your work - simple enough even I could follow!

[dbluhm]

Note to self, this resolver should have a clean up routine run when the associated connection record is deleted, similar to did:peer:2/3

Took a closer look at this. This is a problem that actually just was never solved in the legacy peer case as well. I think fixing this belongs in a future PR. I'll log an issue.

[dbluhm]

Issue created: #2654

[Jsyro]

Looks great, one question though.

This maintains existing storage and fetching behaviour of the legacy unqualified DIDDocs, those documents are not going to be converted to did:peer:1?

[swcurran]

This maintains existing storage and fetching behaviour of the legacy unqualified DIDDocs, those documents are not going to be converted to did:peer:1?

No — existing documents need not be converted, since they are only ever used internally, never with external parties. At some point we might update them, although almost certainly not to did:peer:1.