Upgrade Prometheus

[pracucci]

What this PR does:
In this PR I'm upgrading Prometheus to latest main.

The following Prometheus changes affect us:

• Decrement active_appenders metric when no samples added prometheus/prometheus#9230
• Remove symbols map from TSDB head prometheus/prometheus#9301
• Fix div by 0 panic is resize function. prometheus/prometheus#9286
• Fix a data race in the loadWAL function caused by reusing the same error var in multiple go routines. prometheus/prometheus#9259

There are 3 other PRs affecting TSDB, but these should affect only snapshotting which is a feature we're not using in Mimir yet:

• Do not replay checkpoint if it is covered by snapshot prometheus/prometheus#9226
• Exemplars in snapshot prometheus/prometheus#9255
• Fix deletion of old snapshots prometheus/prometheus#9314

This upgrades Alertmanager too, adding SNS receiver. Unfortunately the SNS receiver HTTP client is unused, so I've open an Alertmanager fix PR first (the PR is included in vendor/ changes here, so lint CI is expected to fail).

Other updated vendored libs:

• github.com/aws/aws-sdk-go v1.40.10 from to github.com/aws/aws-sdk-go v1.40.37: checked the CHANGELOG and I can't see any change affecting us
• github.com/hashicorp/consul: checked the diff and changes are not affecting us.
• github.com/hashicorp/memberlist: checked the diff and I can see few fixes and IngestionAwareTransport deprecation (we don't use it).
• cloud.google.com/go from 0.87.0 to 0.92.0: checked the CHANGES, nothing affecting us.

Which issue(s) this PR fixes:
Fixes #192

Checklist

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

[pracucci]

This upgrades Alertmanager too, adding SNS receiver. Unfortunately the SNS receiver HTTP client is unused, so I've open an Alertmanager fix PR first (the PR is included in vendor/ changes here, so lint CI is expected to fail).

The PR has been merged. To include it, I had to upgrade Alertmanager to latest main. Compared to 0.23.0 (which is the version against which I've checked the diff when I've opened the PR), the following changes have been introduced affecting us:

• Secure cluster traffic via mutual TLS prometheus/alertmanager#2237 (should be safe, unless you add TLS support which is experimental so far)
• Enable support for custom callbacks as part of maintenance prometheus/alertmanager#2689 + Correctly call default silence maintenance function prometheus/alertmanager#2701 (allows to override the maintainance function, looks safe after the fix in 2701)
• Opsgenie improvements:
  • Add the option to update message and description when sending alerts to opsgenie prometheus/alertmanager#2519
  • OpsGenie: refer to alert instead of incident prometheus/alertmanager#2609

[pstibrany]

Good job.

[pstibrany]

Let's also add "sns" into allowedIntegrationNames in notifications_limit_flag.go

[pracucci]

Thanks for spotting it!

[bboreham]

lgtm, didn't see anything alarming in vendor changes.