Make `disrupt_mgmt_restore` nemesis independent on DC configuration (if they mismatch for cluster and backup snapshot)

[mikliapko]

The detailed issue description is provided in PR comment where it was observed and in a separate Manager ticket with a request to fix it on Manager side.

I suppose that the only feasible workaround for now is to apply schema restore statement out of Manager as direct CQL calls.

It was also a suggestion from @Michal-Leszczynski to add some flag to Manager which will allow to retrieve the schema restore statement from backup snapshot.

Which way to go will actually depend on how soon we need to have the fix in SCT (scylladb/scylla-cluster-tests#10586 (comment)).