Add retry for cloud storage cleanup

Hi,

It is observed during tests that if backup pod is deleted during backup, the new backup pod is started and created backup files are cleaned up. But after the cleanup the new files can be additionally uploaded to the storage after cleanup and before the actual SST stream starts and as a result backup fails.

We need to add retry logic to cleanup to check that if files were uploaded to storage after the cleanup - we clean them up also.

STR: run demand-backup-cloud test on aks norwayeast or gke west.