add vector benchmark and execute it on cooley, theta, and bebop

See merge request !8

- incomplete at this point
- just a slightly stripped down variation of margo-p2p-bw

- adjusting to conventions used in sds-repo now

While we should be aware of spack changes, let's have the nightly tests
exercise mochi stuff.  We'll update spack when spack releases happen.

regression pkg rename

See merge request !6

remove barriers in bw benchmarks

See merge request !5

- also switch bake benchmark to not use a progress thread

turn off progress thread in margo benchmarks

See merge request !4

the 'modules' commands won't show us what we are looking for now that
spack does all that itself.

yes this is the same as the bebop fix from a few weeks ago. sigh.

I don't know if 100 iterations is enough to give us confidnece in the
numbers, but I had to crank the latency test way way down to get this
test to complete in a reasonable amount of time.

nightly tests have been failing with "pmemobj_create: open
"/dev/shm/mobject.dat": File exists".  Guess an earlier test bombed out
before it could clean up.

- test now shuts down cleanly: no need to 'kill' anything
- found a few problems with the job script

"JOB4.*" will resolve to just .* and probably that enormous email is why I
didn't see anything from theta.

- seems like a somewhat more general alternative to -d 64, but with the
  same net result