DT-1042 - Improve user experience on Failed Snapshot Delete

[snf2ye]

Changes included in PR:

1. Switch to default successful undo instead of throwing fatal error exception on UNDO
   Why?

• We still don't want to 'UNDO' delete operations - By just passing through the undo steps, we don't actually try to undo any delete operations. "UNDO" would actually mean recreating resources and that is way too messy.
• Allow unlock undo steps to run - By passing through the undo steps, we eventually hit the "undo" steps for the snapshot and dataset lock steps. This unlocks the snapshot and dataset so they don't "disappear" in the user's view.

2. Reorder SAM resource delete step
   If we want to re-run the delete flight, we need the SAM resource to still exist. So, we should do this step last so that we can ensure all of the rest of the delete steps passed before removing it.

3. Add note about snapshot delete in open api spec
   I believe the main goal of the confusing error messages previously returned is to alert the user that UNDOing the delete didn't work. I think we can directly communicate this via the endpoint description.

[sonarqubecloud]

Quality Gate passed

Issues
3 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud