[OpenMetricsV2] Improve label sharing behavior

[ofek]

What does this PR do?

1. Change default behavior of share_labels to v1 label_joins i.e. cache previous payloads
2. Add option cache_shared_labels to prevent inconsistences at the risk of potentially increased memory usage

Motivation

Potential inconsistences are preferred over dependence on the order in which metrics are exposed in a payload

[codecov]

Codecov Report

Merging #9804 (bb50d47) into master (47882ad) will decrease coverage by 0.05%.
The diff coverage is 97.72%.

Flag	Coverage Δ
amazon_msk	87.94% <ø> (ø)
ambari	86.98% <ø> (ø)
apache	95.43% <ø> (ø)
btrfs	82.91% <ø> (ø)
cilium	85.84% <ø> (+1.88%)	⬆️
clickhouse	96.55% <ø> (ø)
cloud_foundry_api	95.98% <ø> (+0.12%)	⬆️
cockroachdb	97.18% <ø> (ø)
consul	93.04% <ø> (ø)
coredns	96.36% <ø> (ø)
crio	100.00% <ø> (ø)
datadog_checks_dev	79.91% <ø> (-0.03%)	⬇️
directory	94.70% <ø> (ø)
disk	84.66% <ø> (-6.85%)	⬇️
dotnetclr	100.00% <ø> (ø)
druid	97.70% <ø> (ø)
ecs_fargate	77.65% <ø> (ø)
eks_fargate	94.05% <ø> (ø)
elastic	88.54% <ø> (ø)
etcd	93.09% <ø> (ø)
external_dns	100.00% <ø> (ø)
fluentd	94.77% <ø> (ø)
gearmand	77.27% <ø> (+1.29%)	⬆️
haproxy	95.22% <ø> (+0.17%)	⬆️
hazelcast	92.39% <ø> (ø)
hdfs_namenode	87.94% <ø> (ø)
http_check	89.96% <ø> (+1.82%)	⬆️
iis	92.41% <ø> (ø)
kong	92.21% <ø> (ø)
kube_apiserver_metrics	97.35% <ø> (ø)
kube_controller_manager	97.05% <ø> (ø)
kube_dns	98.85% <ø> (ø)
kube_metrics_server	100.00% <ø> (ø)
kube_proxy	100.00% <ø> (ø)
kubelet	89.47% <ø> (ø)
kubernetes_state	89.69% <ø> (ø)
lighttpd	83.64% <ø> (ø)
linux_proc_extras	96.22% <ø> (ø)
mapr	82.53% <ø> (ø)
mapreduce	82.27% <ø> (ø)
marathon	83.12% <ø> (ø)
mesos_master	89.68% <ø> (ø)
mesos_slave	93.63% <ø> (ø)
nagios	89.53% <ø> (ø)
network	76.61% <ø> (-0.15%)	⬇️
nginx	95.11% <ø> (+0.93%)	⬆️
openldap	96.33% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

[ruthnaebeck]

docs review

[ruthnaebeck]

👍 for docs