refactor component runner arg list assembly to handle lists of inputs

[gknapp1]

Cherry-picked feature from #133.

This simplifies the configure, execute, and postprocess argument handling and adds capability needed to pass list-values as arguments to these steps.

Example:

Myna input file:

steps:
- calibrate_additivefoam_heatsource:
    class: single_track_calibration
    application: additivefoam
    configure:
      experiments: experiments.yaml
      nvalues: [0.0, 2.0, 5.0, 7.0, 9.0]

Becomes this argument list for the configure step, which is correctly handled by argparse, assuming the --nvalues parameter is configured to take multiple entries:

['--experiments', 'experiments.yaml', '--nvalues', '0.0', '2.0', '5.0', '7.0', '9.0']

The previous implementation would have returned the following, which argparse can't handle:

['--experiments', 'experiments.yaml', '--nvalues', '"[0.0, 2.0, 5.0, 7.0, 9.0]"']

[streeve]

So this can be either the list or the individual step now?

[gknapp1]

So this can be either the list or the individual step now?

I added an example in the PR summary to illustrate the purpose of this better.

[gknapp1]

Sorted out the CI failures. Part of the previous CI failures was actually catching a bug in the code. The other part of it was due to oversubscribing during the AdditiveFOAM examples using mpirun. I'll pull out the unrelated CI changes to a new PR.

[gknapp1]

@streeve, I cleaned this up. It should be good to go now.

[gknapp1]

/gemini review

[gemini-code-assist]

Code Review

This pull request refactors the argument list assembly to support list-valued arguments and improve code structure. The introduction of the _get_arglist helper function successfully centralizes argument parsing logic and removes code duplication. However, there is a subtle bug in the implementation of _get_arglist due to the way boolean types are handled in Python, which could lead to incorrect command-line arguments for False values. I've provided a specific comment with a code suggestion to fix this. Overall, this is a good refactoring that improves functionality and maintainability.