Adding printer columns to inference model

[kfswain]

Fixes: #504

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: kfswain

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [kfswain]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[netlify]

✅ Deploy Preview for gateway-api-inference-extension ready!

Name	Link
🔨 Latest commit	6a394df
🔍 Latest deploy log	https://app.netlify.com/sites/gateway-api-inference-extension/deploys/67e43c387275ea000843ec93
😎 Deploy Preview	https://deploy-preview-574--gateway-api-inference-extension.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[smarterclayton]

Criticality is operationally important

[kfswain]

I looked into adding it, but it's optional, so I thought that might be odd to have some empty entries.

For posterity/context: We specifically made this field optional so it could be nullable, should we ever make a replacement field we could retroactively one-of them.

That being said, we can still add it, as I agree, it is important.

[smarterclayton]

Also inference model please (the referent)

[kfswain]

Also inference model please (the referent)

targetModels is an array, which is also optional. So just to talk this out (which is kinda why I omitted), what behavior do we want? Is a count sufficient and then the user describes the model they want to look into further?

That could be useful as:
0 implies that modelName==referentName
1 implies modelName is just an indirection
>1 implies there is a transition state or traffic split of some sort ongoing

LMKWYT

[kfswain]

I get the above output using:

// +kubebuilder:printcolumn:name="Model Name",type=string,JSONPath=`.spec.modelName`
// +kubebuilder:printcolumn:name="Target Models",type=string,JSONPath=`.spec.targetModels[*].name`
// +kubebuilder:printcolumn:name="Inference Pool",type=string,JSONPath=`.spec.poolRef.name`
// +kubebuilder:printcolumn:name="Criticality",type=string,JSONPath=`.spec.criticality`
// +kubebuilder:printcolumn:name="Age",type=date,JSONPath=`.metadata.creationTimestamp`

Its worth noting that criticality is specified on all models, the entry will just be blank if unspecified.

I think criticality is fine as is, target model(s) I think is very misleading, as it only prints the first entry

[ahg-g]

/lgtm

[smarterclayton]

The target model thing seems like a bug, I designed custom printers to support arrays, but something may have changed since then.

[smarterclayton]

You might need to use “range” in the target model column.