Use job scheduling information to allow more flexible job submission

[thomasmfish]

This is to enable me to use the JobScheduler for B24. The main aim of this was for me to be able to manually specify and and run a bunch of jobs in the same scheduler in a way that doesn't rely on slicing (e.g. I am likely to want 3 jobs for different IMOD reconstructions and 1 job for AreTomo reconstructions to be run all in one).

This has some changes that will affect existing usage, so I wanted to start a discussion before going further with the JobController. The basics of this are:

• JobSchedulingInformation class that contains all the job information required for submission, as well as the job status information.
• SchedulerModeInterface is now JobSlicerInterface and the function just accepts slicer_params and an input JobSchedulingInformation and outputs a list of JobSchedulingInformations.
• ProgramWrapper no longer has the set_cores and get_args methods as that is handled by either setting them in the initial JobSchedulingInformation object, or when create_sliced_processing_jobs or create_sliced_aggregating_jobs are called (these output lists of JobSchedulingInformation objects).

To do:

• JobController needs some work, along with run_ppc, and tests for those. This is where I need to know what's easiest for existing projects - perhaps there's some compatibility layer I can add to avoid updating the wrappers?
• I want to make sure all the typing is done in a more standardised way (there's a mix of styles e.g. list[int] and List[int]).

Questions:

• Does all of this sound okay and usable?
• Are there any changes that would make this fit better with current usage?
• What is the best way of moving ahead with JobController? There have been some changes there but the handling of the resource setting via the wrappers has been a point where I'm unsure of how to handle it, especially since I wont be using that part.
• Is this best not being merged but being a fork or something?

[thomasmfish]

All the tests are passing for me too

[PeterC-DLS]

@VictoriaBeilsten-Edmands , could you review this too to update yourself with the changes?

[VictoriaBeilsten-Edmands]

That looks good and makes sense to me.

[thomasmfish]

@PeterC-DLS Please merge this whenever you're ready (I'd ask you to approve but why then wait for me to click a button?)

[PeterC-DLS]

I was going to test with a real submission job but got side-tracked by type checking...

[thomasmfish]

Okay, no worries, I just didn't want you to be waiting on me!

[thomasmfish]

Just pushed a small change where a couple of error messages were missing the f to make it an f-string but otherwise I'm happy with everything

EDIT: Just saw the but about the n/a path, so pending that