Restrict update column in batches for large tables

See merge request gitlab-org/gitlab-ce!15458

add_column_with_default is implemented in terms of update_column_in_batches, but
update_column_in_batches can be used independently. Neither of these should be
used on the specified large tables, because they will cause issues on large
instances like GitLab.com.

This also ignores the cop for all existing migrations, renaming
AddColumnWithDefaultToLargeTable where appropriate.

Remove the selects when counting the last page

Closes #40266

See merge request gitlab-org/gitlab-ce!15456

Add request rate limits

Closes #30053

See merge request gitlab-org/gitlab-ce!14708

Resolve "/groups/gitlab-org/labels.json redirects to /groups/gitlab-org/-/labels/.json"

Closes #40230

See merge request gitlab-org/gitlab-ce!15427

The last page of the first collection is only loaded into memory when
it is being viewed. If it isn't loaded into memory, the `#size` call
triggers a count query.

This `#count` would generate an invalid query if our custom preloaded
counts are included by adding a separate `as count_column` alias on
top of the count aliases. Removing the selects in this case will make
sure a valid `COUNT(*)` is generated.

Fix hashed storage with project transfers to another namespace

Closes gitlab-ee#4056

See merge request gitlab-org/gitlab-ce!15444

Delete empty fork networks

See merge request gitlab-org/gitlab-ce!15373

Optimise getting the pipeline status of commits

See merge request gitlab-org/gitlab-ce!15332