[RKE2] S3 snapshot retention wording

[nwmac]

Internal Reference: SURE-9250

The working of the retention settings for RKE2 and S3 is not accurate and needs to be reviewed and updated to be clear on what the settings actually mean.

[gaktive]

Some details:

With the current retention logic working as intended, the dashboard appears to add confusion with the "snapshots per node" wording. As the most basic solution to this issue relates to a discrepancy in wording in the UI/documentation, raising this as a bug.

The logic for retention applies to all files in S3, not the number of files per node. With multiple etcd nodes, the retention needs to account for the number of nodes.

rancher/rke2#5216
rancher/rancher#43872

The core issue to fix in the short term is the UI needs to update the wording that the number of snapshots taken is across ALL nodes.