[receiver/k8sclusterreceiver]: add k8s.pod.ready metric

[gdvalle]

Description:
Introduce a new metric, k8s.pod.ready, to go alongside k8s.container.ready. #32941

This is helpful to represent, as k8s.container.ready alone doesn't tell the full picture of whether a pod's ready for service, due to pod ReadinessGates.

With this, we don't need to rely on kube-state-metrics for the info.

This does the simple thing and adds an analog for the pod ready condition, but I could also see a more generalized version, i.e. k8s.pod.status.condition, where we could capture all conditions (API docs), e.g. 0=False, 1=True, 2=Unknown, along with reason in attributes. Maybe message too, if cardinality potential isn't a concern here.

Testing: A couple unit tests were added to assert ready pod condition is correctly recorded for both true and false states.

Documentation: Only auto-generated docs.

[povilasv]

Hey, please create an issue, attach the issue to description and changelog entry.

[gdvalle]

Hey, please create an issue, attach the issue to description and changelog entry.

Done, see #32941

[github-actions]

This PR was marked stale due to lack of activity. It will be closed in 14 days.

[gdvalle]

not stale

[povilasv]

I think maybe this should be disabled by default?

I'm a bit worried It would add quite a bit of new metrics 🤔

[gdvalle]

I'd argue that pod ready state is pretty fundamental to k8s, and it should be part of the default set. If k8s.container.ready is, then this should be too. I'd think for most workloads pod ready state is more important at a high level than container, as it affects whether a Service directs traffic to it. It's also part of the default set in KSM.

That said, I'm happy to flip it if that's what it takes to get this over the line.