Before, compute_ctl didn't have a good registry for what command would run
when, depending exclusively on sync code to apply changes. When users have
many databases/roles to manage, this step can take a substantial amount of
time, breaking assumptions about low (re)start times in other systems.

This commit reduces the time compute_ctl takes to restart when changes must
be applied, by making all commands more or less blind writes, and applying
these commands in an asynchronous context, only waiting for completion once
we know the commands have all been sent.

Additionally, this reduces time spent by batching per-database operations
where previously we would create a new SQL connection for every user-database
operation we planned to execute.

The previous iteration always connected to every database, which slowed down
startup when most databases had no operations scheduled.

Addresses @ tristan957's comment, and allows separate review of these queries.

Instead of Mutex, using an RwLock on shared modifyable state helps speed
up read-only parallel steps (i.e. most parallel steps). This makes the
per-database jobs actually work in parallel, rather than serialize on
accessing the database- and roles tables.

Additionally, this patch reduces the number of parallel workers to 1/3rd of
max_connections, from max_connections - 10. With the previous setting I
consistently hit issues with "ERROR: no unpinned buffers available", now
it works like a charm.