Add skeleton BBH pipeline

[nilsvu]

Proposed changes

Tools to generate-id > start-inspiral > start-ringdown. No automation yet and barely any smart parameter choices, but this should provide a good starting point to flesh out the transition from inspiral to ringdown. See #5021 for details on future plans.

Try it like this:

./bin/spectre bbh generate-id -q 1 -D 20 -w 0.015 -o ./ID
# wait...
./bin/spectre bbh start-inspiral ./ID/InitialData.yaml -O ./Ev/
# wait...
./bin/spectre bbh start-ringdown ./Ev/Inspiral.yaml -O ./Ringdown/

(this won't really work yet because the inspiral won't finish properly, but that's #4600)

Upgrade instructions

Code review checklist

• The code is documented and the documentation renders correctly. Run
  make doc to generate the documentation locally into BUILD_DIR/docs/html.
  Then open index.html.
• The code follows the stylistic and code quality guidelines listed in the
  code review guide.
• The PR lists upgrade instructions and is labeled bugfix or
  new feature if appropriate.

Further comments

[knelli2]

I think this is a great start! We'll definitely need to tune input file parameters and such, but that can be done as we go.

[knelli2]

How do you feel about using another yaml to specify these? As we add more and more capability, having to write a new click option for every nob we want to tweak will probably get pretty annoying. This could simplify the command down to

spectre bbh generate-id InitialParams.yaml

where InitialParams.yaml is

mass-ratio: 1.5
separation: 16
orbital-angular-velocity: 0.01
polynomial-order: 7

and then all the other parameters we didn't specify will be default.

This suggestion applies to all phases of the BBH pipeline.

[nilsvu]

Ohh no, adding another "meta" input file seems like it could be a disaster :D I'd rather keep the knobs here down, and even remove the hp refinement options once we have AMR. If you want full control over options you can write your own input file and run it. The steps in the pipeline still work if you're starting them from a simulation that you didn't run with the pipeline.

[nilsvu]

I would be fine with adding an option to supply the CLI option from a configuration file rather than on the command line if you want. We already do something like that with the user config file and the status CLI. It's just an alternative way of passing the options to the CLI. There's a pretty generic way to do that with click.

[knelli2]

Yeah I think that would be nice.

[knelli2]

Did you want to make this change now or in a follow up PR?

[nilsvu]

I wasn't planning to do this right away, since I don't have a use case right now

[nilsdeppe]

I agree, could be nice, first need to see how all this works in practice :)

[knelli2]

This is a great start!

[knelli2]

Did you want to make this change now or in a follow up PR?

[nilsvu]

Fixed the MPI failure (I hope)

[nilsdeppe]

I didn't read it all in fine detail, but this seems like a good way to structure and organize all this.

[nilsdeppe]

I agree, could be nice, first need to see how all this works in practice :)

[nilsvu]

Rebased