[Story] Pluggable Discovery Microservices

[andrewazores]

Parent: https://github.com/cryostatio/cryostat/issues/936
Depends on https://github.com/cryostatio/cryostat/issues/939
Depends on https://github.com/cryostatio/cryostat/issues/1049

Once we have an API designed for clients to provide environment and target node information to Cryostat, we should implement at least one such client ourselves as an optional microservice to deploy alongside Cryostat. The core functionalities of custom targets, JDP discovery, and OpenShift API (Endpoints and/or Annotations) should be included by default within the Cryostat deployment to minimize the footprint and maintenance burden but should be architected similarly to what we expect microservices to do, except running in-process (but still optional and enabled by environment variable(s)).

• [Task] Document expected behaviour for microservices #1050
• [Task] Implement a small example/template microservice (ex. for JDP) #1051
• (Optional) Extract AbstractNode, EnvironmentNode, TargetNode, ServiceRef, and other relevant useful classes into -core for reuse by microservices implemented in JVM languages. (not strictly necessary, JVM services can avoid using -core so long as the serialized API responses have the correct fields, and same for non-JVM languages. But, for JVM languages, it would be nice to have for reuse to make implementation a little easier). Perhaps rather than -core this should be a separate interface/structs-only dependency to keep it small, ex. cryostat-discovery-api.

[github-actions]

This PR/issue depends on:

• cryostatio/cryostat#939
• cryostatio/cryostat#1049
  By Dependent Issues (🤖). Happy coding!