chore: Update BDRS to upstream 0.15.1

[lgblaumeiser]

WHAT

Update the BDRS repo to the latest upstream version 0.15.1.

WHY

To stay with upstream development
To fix the openapi issue which is based on a mismatch between the autodoc version used and the latest upstream workflows in use for openapi creation.

Closes #325

[lgblaumeiser]

Hi @ndr-brt ,

would be nice, if you could have a look, I hope, I did not miss a thing concerning the update to 0.15.x. The audience thing is a bit difficult, as the token used by a bdrs client contains for all three claims, issue, subject, and audience the callers did, at least for the audience I would have expected the bdrs operators did, but this is not really specified, as it is a heuristic, that the issuers did is also the did of the BDRS operator. I assume that was the reason for that decision. So I faked the audience check, as it uses the value missing some other source for the value.

[ndr-brt]

the audience used to validate shouldn't be the one in the token, otherwise it will always pass, because the validator also extracts the audience from the token for comparison.

The audience passed should be the own did of the bdrs service, so the configured one.
All the parse logic here can be avoided then

[lgblaumeiser]

Problem is, that this is not the case, see the way the token is generated in the connector:

var claims = Map.<String, Object>of(
  JWT_ID, UUID.randomUUID().toString(),
  ISSUER, ownDid,
  SUBJECT, ownDid,
  AUDIENCE, ownDid
);

Source: BdrsClientImpl.java

My hypothesis is, that when the BDRS was created, this part was done in a rush and the setup with a proper DID on the BDRS side was overlooked. If you look at the charts, there is the following comment:

            - name: "EDC_IAM_ISSUER_ID"
              value: "required but not really needed"

A EDC_PARTICIPANT_ID is not even required for the service.

The way, the authentication is handled in BDRS is not the target solution, imho, but it is as it is today, which means, that the audience is not really used properly. That is, why this code is needed.

[ndr-brt]

I'm ok with your solution them