[ws-manager] add a metric to track volume snapshot time

[jenting]

Description

Add a metric to track the volume snapshot time (the time it takes to create the volume snapshot of a workspace content).

Related Issue(s)

Partially #10195 (will have a PR for volume restore time)

How to test

• Bootstrap a new workspace preview environment using the installer from this PR.
• Add Git integration to GitHub.
• Enable the feature flag persistent_volume_claim.
• Create a new workspace, create a large file in the workspace and stop the workspace.

  dd if=/dev/urandom of=1G.random bs=16M count=64

• Port forward metrics port of the ws-manager and look for new metrics.

  kubectl port-forward <ws-manager-pod-name> 9500:9500
  curl -XGET localhost:9500/metrics

• The reported time gitpod_ws_manager_volume_snapshot_seconds_xxx should match the time it took to perform the volume snapshot.

Release Notes

[experimental] add a metric to track volume snapshot time

Documentation

[jenting]

/werft run with-clean-slate-deployment

👍 started the job as gitpod-build-jenting-observability-add-metrics-10195.5
(with .werft/ from main)