Add logging options to Helm chart

[Baarsgaard]

Expose available logging options in the Helm chart.
Useful for people who use structured logging solutions.
I found #1659 and thought it would be a quick fix.

I also found some securityContext options that were not replicated to/from the kustomize configuration, those were aligned along with the logging args.

Replication steps:

# Prepare kind cluster
export KUBECONFIG=~/.kube/kind-grafana-operator
make kind-start

# Install Helm chart with no modifications
helm upgrade -i grafana-operator ./deploy/helm/grafana-operator --version v5.13.0

# Verify default configurations
kubectl get deployments.apps grafana-operator -o yaml | grep -A 10 containers:
# Verify standard/default log output
kubectl logs deployments/grafana-operator

# Install helm chart with overwritten logging values
helm upgrade -i grafana-operator ./deploy/helm/grafana-operator --version v5.13.0 \
  --set logging.encoder=json --set logging.level=debug --set logging.time=rfc3339

# Verify arg changes in deployment
kubectl get deployments.apps grafana-operator -o yaml | grep -A 10 containers:
# Verify level, encoding, and time format have changed in the logs
kubectl logs deployments/grafana-operator

[weisdd]

Thanks for the PR! Configuration for logging is certainly useful, and nice catch around missing options securityContext in kustomize.

I certainly did not expect to leave so many comments on a simple PR, but since you've made many changes in various places (not necessarily related to the initial idea behind the PR), there are more things to point at.

[Baarsgaard]

@weisdd Thanks for the thorough review! I've applied your suggestions.
On your first comment with regards to the Kustomize options and documenting them.
I was not sure which direction to take with that?
Were you thinking adding an example jsonpatch similar to the argocd patches?

patches:
  - target:
      apiVersion: apps/v1
      kind: Deployment
      name: grafana-operator-controller-manager
    patches: |-
      - op: add
        path: /spec/template/spec/containers/0/args/-
        value: -zap-encoder=json

[weisdd]

@Baarsgaard In my view, the way you did it (commented-out section) is good enough for now. In future, it would be handy to just have a separate page in docs where we would list all available configuration options like logging flags or whatever else. It doesn't have to be ArgoCD-specific - in the end, people who use it know enough to patch resources themselves, they just need to know which arguments to pass.

[weisdd]

During today's meeting, we (maintainers) discussed this PR.

• It was decided to remove references to zap logging flags from Kustomize (done). @theSuess will prepare documentation instead;
• I've noticed that the default setting for time did not match the previous format (rfc3339, which is made the default through operator-sdk), that is fixed now.

For the rest, the PR should be good to merge. Thanks for your contribution!