Use client to verify the model inference is ready in E2E test

[kerthcet]

What would you like to be added:

Right now, we only check the service is ready, but we didn't query the service, which is more intuitive.

Why is this needed:

Completion requirements:

This enhancement requires the following artifacts:

• Design doc
• API change
• Docs update

The artifacts should be linked in subsequent comments.

[kerthcet]

/help

[nayihz]

/assign
Maybe need some time to get involved.

[kerthcet]

The assign command only works for org member right now, I ticked an issue here cli/cli#9620, would like to find out the workaround.

[kerthcet]

The assign command only works for org member right now, I ticked an issue here cli/cli#9620, would like to find out the workaround.

Fixed by #291

[nayihz]

My thought is to create a new Pod and use the curl command to send an HTTP request to verify that the inference service is avaliable. WDYT?

[kerthcet]

Since we have access to the kubeconfig, I think we can port-forward the pod via client-go and visit the localhost for api test. But I haven't tried this before, I asked the chatgpt, it says yes. Maybe you can take a try.

[nayihz]

Close it after #310 merged.
/close