improve migration of community subscriptions: set dspace_object_id to community UUID / remove community_id

[saschaszott]

This PR handles community subscriptions managed in the subscription table: it adds the community UUID to column dspace_object_id. This corresponds to the way new community subscriptions in DS CRIS are managed.

Afterward the column community_id in the subscription table is removed. This column was added with DSpace CRIS based on DSpace 1.8.2.2 (https://github.com/4Science/DSpace/blob/dspace-5_x_x-cris/dspace-api/src/main/resources/org/dspace/storage/rdbms/sqlmigration/postgres/V1.8.2.2__DSpaceCRIS-subscription_database_schema.sql).

[saschaszott]

The in-depth analysis in issue #371 has shown that the table subscription does not have column community_id in a clean/fresh installation of DS CRIS 7. Therefore, it is crucial that changes in this PR do not affect clean/fresh installations of DS CRIS 7. In its current form, this PR does not consider this fact. It needs to be extended.

[saschaszott]

To fix the problem mentioned in the comment below it is crucial that the Flyway script V1.8.2.2__DSpaceCRIS-subscription_database_schema.sql which was shipped with DS CRIS 5 (https://github.com/4Science/DSpace/blob/dspace-5_x_x-cris/dspace-api/src/main/resources/org/dspace/storage/rdbms/sqlmigration/postgres/V1.8.2.2__DSpaceCRIS-subscription_database_schema.sql) is added in org/dspace/storage/rdbms/sqlmigration/postgres.

CRIS installations which have already migrated to DS CRIS 7 need to apply the changes manually.

[github-actions]

This pull request has been automatically marked as stale because it has not had
activity in one year. It will be closed in 14 days if no further activity occurs.

Allowing pull requests to close as stale helps us filter out old work that is no longer
relevant and helps developers focus on reviewing current work.

All pull requests closed by this bot act like normal pull requests;
they can be searched for, commented on or reopened at any point.

If these changes are still relevant then please comment and/or rebase your PR based on the
latest DSpace code. This will remove the stale status and notify us to assign a reviewer
for your PR.

[saschaszott]

Has the problem been fixed in the meantime?