fix: correct quotas on prometheus federation HA

[aslafy-z]

When querying a child prometheus that uses federation to gather his metrics, an additional label is added to the metrics (eg: prometheus_replica). When a rollout is done on the parent prometheuses, there could be two metrics for the same time frame. This causes the scalar( function to return NaN as stated in the documentation, https://prometheus.io/docs/prometheus/latest/querying/functions/#scalar. Using scalar(max( instead of scalar( fixes the issue.

replaces #855

[github-actions]

This PR has been automatically marked as stale because it has not
had any activity in the past 30 days.

The next time this stale check runs, the stale label will be
removed if there is new activity. The issue will be closed in 7
days if there is no new activity.

Thank you for your contributions!

[github-actions]

This PR has been automatically marked as stale because it has not
had any activity in the past 30 days.

The next time this stale check runs, the stale label will be
removed if there is new activity. The issue will be closed in 7
days if there is no new activity.

Thank you for your contributions!

[aslafy-z]

I will resolve conflicts asap

[aslafy-z]

@povilasv Can you please give a look? Thank you