RO-5068 Refactor IDMapper service to use new groups collections

[ChunxiaoB]

• The group_metadata collection contains the content previously stored in four separate group collections.
• Added filtering by group_provenance_id to preserve the existing data retrieval flow and behavior.
• Introduced a centralized bi-directional mapping between aggregation_method and group_provenance_id.
• The IDMapper client continues to use aggregation_method as input. The IDMapper service translates the input to group_provenance_id for getting the results.
• Centralizing the mapping logic eliminates scattered hard-coded enum dependencies and simplifies future maintenance.

The PR passed all Postman tests.

[josemduarte]

LGTM thanks, as long as all other tests look good. Now it will just need adapting to the new split group_metadata collections.