MPCD reverse nonequilibrium shear flow

[wkdarko]

Description

This PR introduces an MPCD particle updater to perform reverse nonequilibrium shear flow simulations using the Müller-Plathe method. The method is related to md.update.ReversePerturbationFlow, which only operates on HOOMD particles, but the implementation here follows the "standard" approach:

• Original paper: https://doi.org/10.1103/PhysRevE.59.4894
• Extension to use a target swap momentum: https://doi.org/10.1063/1.3276454

We had some old comments / discussion related to why the MD implementation may not be optimal for users that we missed putting into an issue:

• Convert Mueller plathe flow to v3 API #946 (comment)
• Query Regarding Specifying Time Steps (W) for Viscosity Measurement Using hoomd.md.update.ReversePerturbationFlow #1682

Motivation and context

This updater allows users to simulate periodic shear flows with MPCD particles and to extract the viscosity from those simulations.

How has this been tested?

Unit tests were added for mpcd.update.ReverseNonequilibriumShearFlow.

Checklist:

• I have reviewed the Contributor Guidelines.
• I agree with the terms of the HOOMD-blue Contributor Agreement.
• My name is on the list of contributors (sphinx-doc/credits.rst) in the pull request source branch.
• I have summarized these changes in CHANGELOG.rst following the established format.

[joaander]

How is this related to the existing ReversePerturbationFlow?

[mphoward]

@joaander I updated the PR description to answer your question. In addition to what's above, longer term, I would favor deprecating the current non-standard implementation of the MD updater to use the same methods that are implemented here for the MPCD particles. This could be done by refactoring this class to operate on both types of particle data.

[joaander]

I would also favor deprecating the current implementation and adding a replacement. If you think it is appropriate and you or your students have time, those changes would be welcome in this PR. Pushing those changes to a later PR is also fine with me - in that case we should open an issue to track our plans.

[mphoward]

I think it will take us a bit of time to do that refactoring, so let's defer the actual work to a future PR. We can mark the existing MD updater as deprecated now though, if you can let me know what steps you are using to document that. I can then also open an issue for the planned work.

[mphoward]

Thanks Kwabena! You need to get the doc build working, and I have a few small comments that need to be addressed. Please message me if you have questions about any of them.

[mphoward]

All my comments have been addressed, and I inspected the documentation. Good to go from my side!

[joaander]

Thanks! There are a few small updates needed to finish this.

[joaander]

Thanks!