Add CFL Based Adaptive Time-Stepping

[wilfonba]

Description

This PR adds constant CFL time-stepping for all physics. The new feature is compatible with the existing ability to restart cases with additional patches.

Fixes #298

Type of change

• New feature (non-breaking change which adds functionality)

Scope

• This PR comprises a set of related changes with a common goal

How Has This Been Tested?

• Test A: This test is a 1D sodshock tube ran with gfortran 14.1.0 on an Apple M2 processor. The figure shows the results of a constant CFL simulation overlayed over a constant dt simulation.
  
• Test B: This test is a 2D Riemann problem ran with gfortran 14.1.0 on an Apple M2 procesor. The video shows the constant CFL result on the left and the constant dt simulation on the right. Some minor differences are expected.
  https://github.com/MFlowCode/MFC/assets/48168887/95bc7182-f013-49c5-8951-960ed68574ea
• Test C: This test shows a simulation that is restarted with a new patch. It was ran with gfortran 14.1.0 on an Apple M2 processor.
  https://github.com/MFlowCode/MFC/assets/48168887/fad28d4f-bec4-455e-a6a8-795db5724f36
• Test D: This test is a 3D shock droplet ran with NVHPC 22.11 on Pace Phoenix with Nvidia V100 GPUs. The left panel shows the constant CFL result and the right panel shows the constant dt result.
  https://github.com/MFlowCode/MFC/assets/48168887/882cf2bc-a5dc-4ef5-a25e-400c3ac3016d

Checklist

• I have added comments for the new code
• I added Doxygen docstrings to the new code
• I have made corresponding changes to the documentation (docs/)
• I have added regression tests to the test suite so that people can verify in the future that the feature is behaving as expected
• I have added example cases in examples/ that demonstrate my new feature performing as expected.
  They run to completion and demonstrate "interesting physics"
• I ran ./mfc.sh format before committing my code
• New and existing tests pass locally with my changes, including with GPU capability enabled (both NVIDIA hardware with NVHPC compilers and AMD hardware with CRAY compilers) and disabled
• This PR does not introduce any repeated code (it follows the DRY principle)
• I cannot think of a way to condense this code and reduce any introduced additional line count

If your code changes any code source files (anything in src/simulation)

To make sure the code is performing as expected on GPU devices, I have:

• Checked that the code compiles using NVHPC compilers
• Checked that the code compiles using CRAY compilers
• Ran the code on either V100, A100, or H100 GPUs and ensured the new feature performed as expected (the GPU results match the CPU results)
• Ran the code on MI200+ GPUs and ensure the new features performed as expected (the GPU results match the CPU results)
• Compared ns/gp/eq/rhs
  V100 - cfl_adap_dt (3.320) constant dt (3.311) 8M Cells on 1 V100
  M2 - cfl_adap_dt (66.645) constant dt (65.657) 1M Cells on 1 core

[codecov]

Codecov Report

Attention: Patch coverage is 79.12088% with 38 lines in your changes missing coverage. Please review.

Project coverage is 54.69%. Comparing base (dec9171) to head (2357359).
Report is 9 commits behind head on master.

Files with missing lines	Patch %	Lines
src/simulation/m_start_up.fpp	76.08%	8 Missing and 3 partials ⚠️
src/simulation/m_time_steppers.fpp	61.53%	8 Missing and 2 partials ⚠️
src/simulation/m_sim_helpers.f90	90.00%	3 Missing and 1 partial ⚠️
src/simulation/p_main.fpp	80.00%	1 Missing and 3 partials ⚠️
src/pre_process/m_data_output.fpp	57.14%	3 Missing ⚠️
src/pre_process/m_start_up.fpp	25.00%	3 Missing ⚠️
src/post_process/p_main.fpp	85.71%	1 Missing and 1 partial ⚠️
src/simulation/m_rhs.fpp	66.66%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #515      +/-   ##
==========================================
+ Coverage   53.99%   54.69%   +0.70%     
==========================================
  Files          58       59       +1     
  Lines       13719    13662      -57     
  Branches     1635     1698      +63     
==========================================
+ Hits         7407     7473      +66     
+ Misses       5867     5740     -127     
- Partials      445      449       +4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[sbryngelson]

@wilfonba this codecov thing is amazing. It shows what lines you added that aren't covered in the test suite! Not that you have to cover 100% of the new lines, but you might be able to see what cases aren't covered that you could easily add.

[wilfonba]

Yeah. I'm sure I can up the coverage a decent amount without too many additional tests. Apparently I need to fix GPUs too

[sbryngelson]

Can you run an ./mfc.sh test -l and copy/paste the new test case [names] here so I know what they are?

[wilfonba]

  0501B3DA   1D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  AE02324F   1D -> Bodyforces -> cfl_adap_dt=T                                      
  D80F2162   2D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  121D4ECA   2D -> Bodyforces -> cfl_adap_dt=T                                      
  128954AD   3D -> Cylindrical -> cfl_adap_dt=T                                     
  07C54EDD   3D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  B54BB9D8   3D -> Bodyforces -> cfl_adap_dt=T 
  37FA2CEF   1D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T  
  6B1AD553   2D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T  
  53A15FFC   3D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T

[wilfonba]

This list was created with #518 . I figured it would be more clear this way.

[sbryngelson]

  0501B3DA   1D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  AE02324F   1D -> Bodyforces -> cfl_adap_dt=T                                      
  D80F2162   2D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  121D4ECA   2D -> Bodyforces -> cfl_adap_dt=T                                      
  128954AD   3D -> Cylindrical -> cfl_adap_dt=T                                     
  07C54EDD   3D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_adap_dt=T  
  B54BB9D8   3D -> Bodyforces -> cfl_adap_dt=T 
  37FA2CEF   1D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T  
  6B1AD553   2D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T  
  53A15FFC   3D -> Viscosity -> Bubbles -> QBMM -> bubble_model=3 -> cfl_const_dt=T

Perfect, thanks

[sbryngelson]

@wilfonba is this ready for review/merge?

[wilfonba]

Most all of the requested changes have been addressed. I feel that updating m_riemann_solvers would be better suited for a separate PR to adhere to the "This PR comprises a set of related changes with a common goal" checkbox. I've already kind of exceeded this with my changes to m_data_output, so I'd like to minimize the violation.

[sbryngelson]

Most all of the requested changes have been addressed. I feel that updating m_riemann_solvers would be better suited for a separate PR to adhere to the "This PR comprises a set of related changes with a common goal" checkbox. I've already kind of exceeded this with my changes to m_data_output, so I'd like to minimize the violation.

Please open an issue detailing what you didn't do, then, please.