Make MPCD collision methods compatible with rigid bodies

[mzbush]

Description

This PR implements a means of momentum transfer from constituent particles to the rigid body during the collision step. This method requires that the constituent particles be thermalized prior to the collision step, where all constituent particles of rigid bodies have a random velocity vector added to their current velocity such that the net gain of these random velocities to the linear and angular velocity of the rigid body is zero. This is to ensure that rigid bodies reach the set temperature in the MPCD fluid.

This implementation is currently incompatible with MPI and only works for 1 CPU or 1 GPU.

Motivation and context

Resolves #2001

How has this been tested?

Unit tests have been added to ensure momentum conservation and error checking.

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.

[mphoward]

Looks great, thank you!

@joaander we have done thorough physical validation of this approach (manuscript being finalized), and we get excellent agreement with state of the art. I would recommend to squash merge this one to due to the commit history.

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[mphoward]

@joaander could you resolve the conflict in the changelog entry based on your release plans? We have continued to use this branch and not found any problems, so we'd like to get it merged soon if possible.

[joaander]

Thanks! And sorry for the delay.

[mphoward]

No problem! I fixed a typo in the CHANGELOG link I noticed.