Feature request: support kstatus based health checking of HTTPProxies

**Please describe the problem you have**

Add support for clients to monitor health of `HTTPProxies` using heuristics defined by the kstatus library https://github.com/kubernetes-sigs/cli-utils/tree/master/pkg/kstatus

This is used by (among others)
* Helm ([link](https://helm.sh/community/hips/hip-0022/), through [flux fork](https://github.com/fluxcd/cli-utils) of `cli-utils`)
* FluxCD ([link](https://fluxcd.io/flux/components/kustomize/kustomizations/#health-checks))
* kpt

Originally submitted by @fergian94 as PR #7648
> Problem
> Since the introduction of kstatus in Helm 4, custom resources can now be monitored during install. For HTTPProxy and ExtensionService CRs, the Valid condition type is not recognized by kstatus logic,
which specifically looks for a Ready condition. This causes Helm to assume resources are immediately ready, defeating the purpose of --wait.

Start with PR #7648 and the HTTPProxy and ExtensionService status handling described in the issue. Compare their Valid condition with the Ready condition expected by the kstatus library, then verify that Helm, FluxCD, and other kstatus-based clients no longer treat these resources as immediately ready.

Written by the indexing model from the issue text.