Build with CUDARTLIB=cudart_static to remove dynamic linkage

Also removed unused curand and nvToolsExt dependencies

BUG 95

Add option to statically link cudart

* Added "verifiable", a suite of kernels for generating and verifying reduction
  input and output arrays in a bit-precise way.
* Data corruption errors now reported in number of wrong elements instead of max
  deviation.
* Use ncclGetLastError.
* Don't run hypercube on non-powers of 2 ranks.
* Fix to hypercube data verification.
* Use "thread local" as the defaut CUDA capture mode.
* Replaced pthread_yield -> sched_yield()
* Bugfix to the cpu-side barrier/allreduce implementations.

as relative to top-level directory. This done is by abspath'ing it before
passing it to subdirectory Makefile's.

The old behavior had two cases: with and without BUILDDIR being set by
the user. With BUILDDIR not set, the build dir would be named "build"
in the top-level directory. If BUILDDIR was set, then the build dir
would be placed at "src/${BUILDDIR}".

The new behavior is simpler, if BUILDDIR is not set then it defaults
to "build", and the directory holding the final build is always at just
"${BUILDDIR}" in the top level.

AlltoAll does not support in-place buffers

ncclGetLastError() was added in NCCL 2.13.0

all files compile now.
mpi tests also pass

Topic/v2.13.4 sync

error introduced with the web merger-resolution tool :-(

fix a messing endif statement

add the rccl/lib directory to the link path

the subdir entry is not actually required for the compilation.

make cmake stage also pass in CI

avoid a division by zero which seems to only occur for op=prod and
datatype=half, since the maximum exponent is small (15) and can exceed
the number of ranks.

fix algorithm assigning values in testsuite

auto-detect and enable MPI

* Adding -pthread flag for linking issues into src/Makefile

* Adding -pthread flag for linking issues into CMakeLists.txt

we honor user requested MPI installations using MPI_PATH first,
and check afterwards for MPICH and Open MPI in the default
Ubuntu and RHEL installation directories.

revamp cmake MPI detection