Use deterministic names for vulnerability reports

[danielpacak]

In the initial release of Starboard we used UUID for different reasons. However, we now believe that Starboard should store the latest vulnerability report for the given workload, and therefore we could use deterministic names.

For example, instead of:

$ kubectl get vulnerabilities.aquasecurity.github.io
NAME                                   REPOSITORY      TAG    SCANNER
6b7b740b-d33c-4f8f-8f0b-af2e94ab1469   library/nginx   1.16   Trivy

$ kubectl tree deploy nginx
NAMESPACE  NAME                                                  READY  REASON  AGE
default    Deployment/nginx                                      -              24h
default    ├─ReplicaSet/nginx-7ff78f74b9                         -              24h
default    │ └─Pod/nginx-7ff78f74b9-pfr6p                        True           24h
default    └─Vulnerability/6b7b740b-d33c-4f8f-8f0b-af2e94ab1469  -              4h2m

We could have:

$ kubectl get vulnerabilities.aquasecurity.github.io
NAME                                   REPOSITORY      TAG    SCANNER
nginx-nginx   library/nginx   1.16   Trivy

$ kubectl tree deploy nginx
NAMESPACE  NAME                                                  READY  REASON  AGE
default    Deployment/nginx                                      -              24h
default    ├─ReplicaSet/nginx-7ff78f74b9                         -              24h
default    │ └─Pod/nginx-7ff78f74b9-pfr6p                        True           24h
default    └─Vulnerability/deploy-nginx-nginx                           -              4h2m

The challenge here is to come up with a naming convention for vulnerability reports to support all types of Kubernetes workloads.

/cc @danielsagi @lizrice @itaysk

[danielpacak]

This has been resolved and we're using deterministic names for all kinds of reports.