Downgrade checks to a warning and not a hard fail

We make certain hardware / base os design decisions, one of them is swap - we don't want it on any of our machines. We order more RAM if we need more memory and we don't do swap or we do it relatively small. This failed on me today:

root@c01:~/backups# skale node restore /root/backups/skale-node-backup-2023-05-22-03-39-16.tar.gz /root/.env
2023-05-23 12:05:22,719 - INFO - Adding BACKUP_RUN to env ...
2023-05-23 12:05:22,720 - INFO - Unpacking backup archive...
2023-05-23 12:05:25,377 - INFO - Executing checks. Type: CheckType.PREINSTALL
2023-05-23 12:05:26,113 - INFO - Host is not fully meet the requirements
==================== Failed checks ====================
+-------+---------------------------------------------+
| Check |                    Info                     |
+=======+=============================================+
| swap  | Expected swap memory 14.9 GB, actual 1.0 GB |
+-------+---------------------------------------------+
Command failed with following errors:
--------------------------------------------------
Restore operation failed
--------------------------------------------------
You can find more info in /root/.skale/.skale-cli-log/debug-node-cli.log

This is against our own design decisions of infrastructure and there is no way to workaround it via cli option or env-file settings.

Start at the restore command's PREINSTALL checks and locate the swap check that reports the expected and actual memory values. Confirm the desired behavior for insufficient swap, then make the check produce a warning rather than aborting restore, and verify that the restore operation continues.