Stabilize experimental corruption check

[gyuho]

We need to stabilize corruption check features with more documentation.

• etcd --experimental-initial-corrupt-check
• etcd --experimental-corrupt-check-time

Reference #9064

[jpbetz]

mTLS required for this to be promoted to be stable

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed after 21 days if no further activity occurs. Thank you for your contributions.

[serathius]

Motivation of reopen: #13775

[lavalamp]

Can we make checking a merkle root part of the graduation criteria? See discussion in: #13839

[serathius]

@lavalamp Depending on the difficulty it might make more sense to graduate corruption check earlier. As https://github.com/etcd-io/etcd/blob/main/Documentation/postmortems/v3.5-data-inconsistency.md#detection states corruption check is currently very limited and there are simpler improvements that we can implemented before Merkle Root.

My current plan would be for v3.6 to make the corruption check reliable by storing hash values in WAL. I would expect Merkle root to be no earlier than v3.7.