GROMACS with CUDA-aware MPI direct GPU communication support

This source code contains the GROMACS version used for the following manuscript: Páll, S., Zhmurov, A., Bauer, P., Abraham, M., Lundborg, M., Gray, A., Hess, B, & Lindahl, E.. (2020). Heterogeneous Parallelization and Acceleration of Molecular Dynamics Simulations in GROMACS. The Journal of Chemical Physics, 2020 The source code is based on the GROMACS 2021-dev code-base version 5a6ece397e36365b2b377ec4b7dbf57265b648c3 the code modifications add CUDA-aware MPI support for halo-exchange and PME-PP communication as well as resolve a number of related issues and limitations. Original git hash of the code is b60021d95ebd1025ff6e235cbb381585d73533d3