Add RayClusterReplicaSet initial implementation

[Jeffwan]

Pull Request Description

RayCluster RS controller implementation. I have not tested it from e2e yet, will cover it later.

Related Issues

Resolves: part of #161

Important: Before submitting, please complete the description above and review the checklist below.

---

Contribution Guidelines (Expand for Details)

We appreciate your contribution to aibrix! To ensure a smooth review process and maintain high code quality, please adhere to the following guidelines:

Pull Request Title Format

Your PR title should start with one of these prefixes to indicate the nature of the change:

• [Bug]: Corrections to existing functionality
• [CI]: Changes to build process or CI pipeline
• [Docs]: Updates or additions to documentation
• [API]: Modifications to aibrix's API or interface
• [CLI]: Changes or additions to the Command Line Interface
• [Misc]: For changes not covered above (use sparingly)

Note: For changes spanning multiple categories, use multiple prefixes in order of importance.

Submission Checklist

• PR title includes appropriate prefix(es)
• Changes are clearly explained in the PR description
• New and existing tests pass successfully
• Code adheres to project style and best practices
• Documentation updated to reflect changes (if applicable)
• Thorough testing completed, no regressions introduced

By submitting this PR, you confirm that you've read these guidelines and your changes align with the project's contribution standards.

[Jeffwan]

@Yicheng-Lu-llll Please take a look at the RayCluster status part. Since RayCluster doesn't have condition, some of the feature in upstream ReplicaSet may not be transferred here. for example, we do not know the RayCluster ready time, correct me if I am wrong.

[Yicheng-Lu-llll]

@Yicheng-Lu-llll Please take a look at the RayCluster status part. Since RayCluster doesn't have condition, some of the feature in upstream ReplicaSet may not be transferred here. for example, we do not know the RayCluster ready time, correct me if I am wrong.

RayCluster now has the condition! Documentation for the condition has just been merged: ray-project/ray#47462. It is available in the latest version of KubeRay.

[Yicheng-Lu-llll]

We can use RayClusterProvisioned to know the first time all Ray pods are ready. After that, depending on how we interpret Ready, we can rely on condition HeadPodReady and ReadyWorkerReplicas in status.

[Jeffwan]

@Yicheng-Lu-llll BTW, is there a stable version release with the condition support? or it's in the master version now?

[Yicheng-Lu-llll]

@Yicheng-Lu-llll BTW, is there a stable version release with the condition support? or it's in the master version now?

v1.2.1 release has the condition support but the default feature gate for this is false. We may need to enable it:

helm upgrade kuberay-operator kuberay/kuberay-operator --version 1.2.1 \
  --set featureGates\[0\].name=RayClusterStatusConditions \
  --set featureGates\[0\].enabled=true

[Jeffwan]

We can use ray-project/kuberay#2301 to know the first time all Ray pods are ready. After that, depending on how we interpret Ready, we can rely on condition ray-project/kuberay#2261 and ReadyWorkerReplicas in status.

I will leave this as a separate issue. tracked here #171