Add triclinic neighbor list support to Kokkos

Fix GPU sync bugs

see commit 09fc8b0b for details on why

because the verlet_kokkos system has
a "clever" optimization which will
alter the datamasks before calling sync/modify,
so the datamask framework must be
strictly obeyed for GPU correctness.
(the optimization is to concurrently
compute forces on the host and GPU,
and add them up at the end of an iteration.
calling your own sync will overwrite
the partial GPU forces with the
partial host forces).

Kokkos bugfixes

fix USER-OMP bug on per-atom data with hybrid styles

warning fixes (Kokkos+CUDA)

Fix skip with ghost issue

USER-DPD: Possible uninitialized variable in pair_exp6_rx.cpp bugfix.

this structure gets put inside a DualView,
so these members need to be able to execute
on the GPU

the compiler pointed out that
boundary[2][2] doesn't exist.
If I understand this correctly,
those checks should be against
boundary[*][0].

this #ifdef adds a return statement
for little endian machines, but leaves
the old one, which the compiler comlains
is unreachable. this commit combines
the conditionals so we can use #else

Added explicit initialization (to zero) for several variables inside the
inner j-loop to avoid using them uninitialized or from prior iterations
within rmOldij_12 == 0.

collected fixes and improvements

Kokkos fixes

USER-DPD another zero compute optimization

This relates to commit 4eb08a58 that was applied to pair_dpd_fdt_energy