Deprecated via https://github.com/coredns/coredns/pull/4781

Drop deprecated metric to fix failing test.

Dropped via https://github.com/kubernetes/kubernetes/pull/103793

Signed-off-by: paulfantom <pawel@krupa.net.pl>

The current implementation of prometheus-adapter exposes a lot of
metrics about the health of its aggregated apiserver. The issue is that
the some of these metrics are not very useful in the context of
prometheus-adapter, and we currently can't avoid exposing them since
they are registered to the Kubernetes global Prometheus registry. Until
this is improved in upstream Kubernetes, we could benefit from dropping
some of the metrics that are not very useful.

Before this change, in a default kube-prometheus installation, we would
have 800+ series for prometheus-adapter against 400+, so we divided the
number of series by two will focusing on the most valuable metrics for
prometheus-adapter.

Signed-off-by: Damien Grisonnet <dgrisonn@redhat.com>

In version [node-exporter v1.2.0](https://github.com/prometheus/node_exporter/releases/tag/v1.2.0

)
two argument name changes were introduced. While the old names still
work (with a deprecation warning), lets use the new names.

Signed-off-by: Jan Fajerski <jfajersk@redhat.com>

This change drops pod-centric metrics without a non-empty 'container' label.

Previously we dropped pod-centric metrics without a (pod, namespace) label set
however these can be critical for debugging.