Some Geo Sidekiq queues not getting drained

From /admin/background_jobs:

In the Sidekiq config in /opt/gitlab/sv/sidekiq-cluster/run, I see:

exec chpst -e /opt/gitlab/etc/gitlab-rails/env -P \
  -U git -u git \
  /usr/bin/env \
    prometheus_multiproc_dir="${prometheus_run_dir}" \
    /opt/gitlab/embedded/service/gitlab-rails/bin/sidekiq-cluster \
      -e production \
      -r /opt/gitlab/embedded/service/gitlab-rails \
        admin_emails,authorized_projects,delete_merged_branches,delete_user,elastic_commit_indexer,elastic_indexer,email_receiver,emails_on_push,export_csv,geo,geo_repository_update,gitlab_shell,group_destroy,ldap_group_sync,mailers,merge,namespaceless_project_destroy,new_issue,new_merge_request,new_note,pages,post_receive,process_commit,project_cache,project_destroy,project_export,project_service,project_update_repository_storage,propagate_service_template,reactive_caching,repository_fork,repository_import,system_hook_push,update_merge_requests,web_hook \
      --negate \

# Do not remove this line; it prevents trouble with the trailing backslashes above.

Some things to point out:

1. Since we're using the --negate flag, in theory we should be picking up everything NOT in that list. However, that doesn't appear to be picking up some queues.
2. geo and geo_repository_update are obsoleted names. I think we can remove them.

/cc: @ahanselka, @jarv