Add persistence to trivy plugin for caching databases

[matheusfm]

Description

This PR adds persistence to trivy plugin for caching vulnerability databases.

• When Zora Helm release is installed with default configuration, a PVC and a Job is created.
• The Job, which uses the PVC, downloads the vulnerability database.
• This Job is required when the persistence is enabled (true by default) so that PVC isn't in Pending status (WaitForFirstConsumer). Since the PVC is created by Helm with --wait flag, this Pending status would be a problem.
• As a best practice, a PV is not created. Just a PVC. Then Kubernetes try to dynamically provision a volume specially for the PVC (Dynamic Provisioning).
• The Java vulnerability database is not downloaded by default (not every cluster has Java apps). But it can be downloaded by setting scan.plugins.trivy.persistence.downloadJavaDB to true.
• The Job is deleted immediately after it finishes, due to ttlSecondsAfterFinished: 0.
• The scheduled Trivy scans use the PVC as well, so the vulnerability databases is just downloaded when needed and no longer for every scan.

How has this been tested?

• kind create cluster
• make docker-build
• kind load docker-image controller:latest
• helm upgrade --install zora charts/zora/ -n zora-system --create-namespace --wait -f values.yaml (content of values.yaml below)
• Check the Job logs: kubectl logs job/trivy-download-db -n zora-system then you should see a Downloading DB... entry
• Check the logs of the 1st Trivy scheduled scan: kubectl logs kind-kind-vuln-trivy-28535863-tdlxz -n zora-system then you should see a DB update was skipped because the local DB is the latest entry.

values.yaml content:

clusterName: kind-kind
operator:
  image:
    repository: controller
    tag: latest
scan:
  plugins:
    trivy:
      persistence:
        enabled: true

Checklist

• I have labeled this PR with the relevant Type labels
• I have documented my code (if applicable)
• My changes are covered by tests