The `#as_indexed_json` method is used to convert individual documents to
the JSON that is sent to Elasticsearch during indexing. We need to
preload the data before it gets to this method. We implemented the
`.preload_indexing_data` method in
https://gitlab.com/gitlab-org/gitlab/-/merge_requests/56808 so we just
need to add this to the various classes now. This MR updates it for
`Issue` only.

We also slightly refactor the issue_instance_proxy since it's more
efficient to just use the `user_id` from the `issue_assignees` join
table than to join again to `users` and also since it wasn't possible to
get the preload to work without loading all of `users.*` which would be
wasteful for a wide table and large numbers of records.