fix: Issue #1391 - Custom labels not being pulled in

[rxraghu]

Why are these changes needed?

fixes #1391 Some organizations have strict labeling requirements for their K8s objects, mainly deployments and pods. kuberay-operator was allowing to add custom labels via values.yaml to pods but not to deployments and kuberay-apiserver was not allowing to add custom labels to either. Added code to pull labels for deployment if present in values.yaml. Also, added code to pull annotations for apiserver as well as is the case with operator.

Related issue number

Closes #1391

Checks

• I've made sure the tests are passing.
• Testing Strategy
  • Unit tests
  • Manual tests
  • This PR is not tested :(

[rxraghu]

I tested this test_sample_rayservice_yamls.py it seems to be failing even without my changes.

[kevin85421]

LGTM. Would you mind adding more details and a screenshot demonstrating how you tested this PR manually before I merge it? There is no test for this PR. Typically, I would clone your fork and conduct manual tests myself, but I have been very busy with the Ray Summit recently.

[rxraghu]

Thanks for the review @kevin85421. Here is how I tested the PR. I added the labels and annotations as shown below to the values.yaml of kuberay-operator and kuberay-apiserver:

labels:
  dummy-label: "dummy-value"

annotations:
  dummy-annotation: "dummy-value"

Then installed both using helm install:

# cd into respective directory
helm install kuberay-operator .
helm install kuberay-apiserver .

Then checked that the dummy label and annotation appeared on the pods and only dummy-label appeared on the deployment using Lens. Attached screenshots for apiserver-deployment, operator-deployment, apiserver-pod and operator-pod.

[rxraghu]

@kevin85421 - Also, can you share when you guys might do the next release? Just to get an estimate of when these changes might be available. Thanks!

[kevin85421]

Also, can you share when you guys might do the next release? Just to get an estimate of when these changes might be available. Thanks!

The Ray Summit is scheduled to take place from September 18 to September 20. Before the Summit commences, we plan to release the first candidate for KubeRay v1.0. Following this, we anticipate a one-month period of iteration with users before officially releasing the final version of v1.0.