Allow specifying queue name & scope of calculation in queue processor

[bdach]

The end goal of this pull request is to be able to split queue processing of beatmaps into two parts: one that populates difficulty attributes exclusively, and one that populates scoring attributes only.

As per usage instructions:

Usage: osu.Server.Queues.BeatmapProcessor [options] <mode> <queue-name>

Arguments:
  mode          The target mode to process the beatmaps from the queue in.
                Allowed values are: All, Difficulty, ScoreAttributes.
                Default value is: All.
  queue-name    The name of the queue to watch. The `osu-queue:` prefix must be omitted.
                Default value is: beatmap.

Options:
  -?|-h|--help  Show help information.

So you'd potentially have two instances of the processor running:

$ dotnet run osu.Server.Queues.BeatmapProcessor Difficulty beatmap-difficulty
$ dotnet run osu.Server.Queues.BeatmapProcessor ScoreAttributes beatmap-score-attributes

correspondingly. The queue-name param feels a little redundant maybe and could be replaced by an implicit hardcoded mapping, but this felt better?

Notably, when the processor is ran without arguments, the defaults are such that it falls back to the current mode of operations with no change in behaviour (everything from osu-queue:beatmap is processed).

I did test this semi-manually on a local environment by pushing items to the queue manually via redis-cli. To make use of the params added here the sending end will have to be adjusted, but I am not sure that is something that I have access to at present.

[smoogipoo]

Your understanding was correct as far as I can see 👍