* added wall param and test

* added wall height parameter and tests for NBody

* fix bug in test construction

* docstring for nbody with new param

* Remove unused variables

* changed wall height param to be a real and modified test to catch the error if it's an int

* modified all test wall params to include one float

* remove dt from dpstokes params

* raul's thrust implementation of position shifting

* modified interface to treat wallHeight param using std::optional

* updated test interfaces to respect new NBody interface: wallHeight only needed when periodicity is set to bottom_wall. if given when periodicity is open, it throws an error

* simplified a test slightly

* interface changes from merge

* slightly simplified position shift in nbody interface

* fix includes to use std library instead of accidentally including through nanobind

* include nanobind versional of optional where the python module gets compiled so it knows how to convert from stl optional to python

* fix bug I introduced in the interface tests

---------

Co-authored-by: Raul P. Pelaez <raulppelaez@gmail.com>

* Remove numberParticles from initialize calls in tests

* Make each solver deal with setting the number of particles

* Remove number of particles in initialization in python wrapper

* Adapt NBody to changes to the number of particles

* Add prints for debug builds

* Fix a file that is supposed to be an install

* Add test for changing the number of particles

* Check that number of particles is not negative

* Formatting

* Add a new function getNumberParticles to the interface.
Remove number of particles from initialize parameters.
Adapt all solvers to be able to take different number of particles
each time.

* Update examples

* Make sure inputs are in CUDA memory

* Add test for thermal drift computation using RFD

* Add thermal drift to the interface, returns 0 by default

* Improve lanczos seeding

* Add a header for RFD

* Update tests

* Add thermal drift to pythonify

* Use RFD in DPStokes

* Make sure linear is filled with 0 in GPU

* Fix bug in random number generation

* formatting

* Error checking and formatting

* Add more tests

* Change RFD delta

* Add thermal drift to NBody

* Small changes to Self mobility

* Change some tolerances

* Allow CPU memory in thermalDrift

* thermal drift should add, not overwrite

* Add thermal drift to hydrodynamic velocities

* comment

* sqrtM should add, not overwrite

* thermal drift should add, not overwrite

* Make sure tensors are zero initialized

* Update documentation

* Update tests/test_thermal_drift.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Fix bug in NBody not correctly substracting the wall height

* Update test

* Update docs

* Update docs

* Update docs

* Update docs

* Add torques to thermal drift interface

* Update comment

* fix variable name

* Add array origins tests for all methods in the interface

* Register device and framework when setting positions

* formatting

* Fix memory location in sqrtMdotW

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Ryker Fish <rfish@mines.edu>

* update tests to check that returns are the size of the positions

* change shapes to by dynamically set based on shape of position

* change how shapes are detected, only allow 2 formats

* add test for bad position shape

* removed silly comment

* update interface to return in shape of input forces/torques when possible and in shape of input positions if not

* update interface tests

* clean up interface/test

* fix typo & rename

* formatting commit only

* change thermal drift shape detection

* fix issue in sqrtM and tests from rebase

* thermal drift test for shapes

* Small change

* Formatting

* Formatting

---------

Co-authored-by: Raul P. Pelaez <raulppelaez@gmail.com>