PuReMD-old: fix communications to only use full shell pattern for non-charge methods (bug only present using neutral territory communication method). TODO: revisit code structure at a later date.

PuReMD-old: fix convergence checks for PIPECG and PIPECR. Update CG to used preconditioned residual in it's check.

PuReMD-old: protect against buffer overflows for traj_title. NOTE: strcpy IS NOT SAFE, DO NOT USE EVER. USE strncpy INSTEAD.

PuReMD-old: back-port run_sim.py from master branch. Update control file parsing to ignore sPuReMD/PG-PuReMD options.

PuReMD-old: optimizations in Init_Bond_Full for fusing hbond and bond list computation and removing unnecessary sorting and sym_index updating.

PuReMD-old: clean-up and comment initialization optimizations. WIP: remove redundant computation of bonds list (not working for full far nbrs list). Improve hbond list computation for full far nbr list case.

PuReMD-old: complete PIPECG and PIPECR implementations. Fix dense linear algebra code to loop beginning with lowest address in memory (vector.c -> level 1 BLAS-like operations).

Merge branch 'full-far-nbrs-list-charge-matrix' of https://gitlab.msu.edu/SParTA/PuReMD into full-far-nbrs-list-charge-matrix.

Merge branch 'full-far-nbrs-list-charge-matrix' of https://gitlab.msu.edu/SParTA/PuReMD into full-far-nbrs-list-charge-matrix

Merge branch 'full-far-nbrs-list-charge-matrix' of https://gitlab.msu.edu/SParTA/PuReMD into full-far-nbrs-list-charge-matrix.

PuReMD: split Init_Forces into 3 functions: Init_Distance, Init_CM and Init_Bond, and add timings for those functions

PuReMD-old: fix timing issues. Remove fine-grained timing of Init_Forces due to high overhead. Disable SAI preconditioner optimization of changing far nbr list and matrix formats based on preconditioner refactoring step (not working). Tweak to not product numeric overflow when setting up SAI preconditioner (pivot sampling).

PuReMD-old: more clean-up of linear algebra code. Remove or conditionally enable debugging code (DEBUG preprocessor). Fix memory leak in SAI setup. Better memory allocation checks.