benchpark python script to grab programming models

[pearce8]

A script to access info in experiment.py:

class Amg2023(
    Experiment,
    OpenMPExperiment,
    CudaExperiment,
    ROCmExperiment,
    StrongScaling,
    WeakScaling,
    ThroughputScaling,
    Caliper,
):

For now, we would use benchpark-python. Once experiments become Spack packages, we could use
spack-python (https://spack-tutorial.readthedocs.io/en/latest/tutorial_spack_scripting.html). #138 is an example of using spack-python.

• benchpark list experiments openmp
• benchpark list experiments cuda
• benchpark list experiments rocm
• benchpark list experiments strong
• benchpark list experiments weak
• benchpark list experiments throughput
• benchpark list experiments caliper
  • caliper is a modifier, so it is listed under benchpark list modifiers
• benchpark list programmingmodels experimentname -> benchpark list experiments --experiment cuda rocm openmp
• benchpark list scaling experimentname -> benchpark list experiments --experiment weak strong
• remove cuda/rocm/openmp from tags in application.py and experiment.py files
• update script generating experiment table to use benchpark list experiments --experiment cuda rocm openmp for the programming-model row
• add scaling-experiments row in experiment table using benchpark list experiments --experiment weak strong command
• add instrumented-caliper row in experiment table using benchpark list modifiers --name caliper --experiments command
• add all new commands to docs

[michaelmckinsey1]

Currently, benchpark list experiments does the following:

$ benchpark list experiments
Experiments:
    ad+single_node
    amg2023+single_node
    amg2023+openmp
    amg2023+cuda
    amg2023+rocm
    amg2023+strong
    amg2023+weak
    amg2023+throughput
    ...

To enable this Issue, would need to add a subparser that down-selects from the current output.