Surface Kubernetes client rate limiter metrics

[barkbay]

This PR helps to surface the latency intentionally added by the Go client in order to throttle the requests ran against the K8S control plane.

The PR basically adds a new Prometheus histogram named k8s_client_rate_limiter_duration_seconds to collect the observed latency, here is an example of the output:

# HELP k8s_client_rate_limiter_duration_seconds Kubernetes client rate limiter latency in seconds. Broken down by verb, and host.
# TYPE k8s_client_rate_limiter_duration_seconds histogram
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="0.005"} 86
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="0.025"} 86
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="0.1"} 87
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="0.25"} 90
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="0.5"} 96
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="1"} 109
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="2"} 114
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="4"} 122
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="8"} 126
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="15"} 127
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="30"} 127
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="60"} 127
k8s_client_rate_limiter_duration_seconds_bucket{host="k8s_host",verb="GET",le="+Inf"} 127
k8s_client_rate_limiter_duration_seconds_sum{host="k8s_host",verb="GET"} 76.791233874
k8s_client_rate_limiter_duration_seconds_count{host="k8s_host",verb="GET"} 127

[thbkrkr]

LGTM