Remove CortexQuerierCapacityFull alert

[pracucci]

What this PR does:
I propose to remove CortexQuerierCapacityFull alert. It was added as part of the initial commit.

Why removing it? My thoughts:

• The alert fires when a single querier is running queries at full capacity (eg. if configured with 4 workers, all workers run queries for 5 consecutive minutes)
• Having a querier running a full capacity is not an issue per se (if we give a querier 4 workers it means we're willing to use all 4 of them if required) and this looks more alerting on cause then symptom
• We already have alerts on queries stuck in query-frontend/scheduler and we have SLO-based alerts, so I'm not really sure this alert is giving us a real value

Which issue(s) this PR fixes:
N/A

Checklist

• CHANGELOG.md updated - the order of entries should be [CHANGE], [FEATURE], [ENHANCEMENT], [BUGFIX]

[gouthamve]

LGTM

[pracucci]

I'm not going to merge it immediately. I want to give other people the time to review and think about it.

[pstibrany]

Your reasoning makes sense to me. 👍

[pracucci]

Going to merge this. I will promptly address any post-merge feedback (if any).