SLURM_JOB_PARTITION to be the one in which the job started.

Bug 1407
Continuation of da8409c0

If using proctrack/cgroup and gres/gpu, always start CUDA_VISIBLE_DEVICES
environment variable numbering at 0.
bug 1421

reservations.

This is an add on to commit 2e5142ef.  Servicing Bug 1418.

the partition in which the job runs.

This reverts commit c18a3c72.

Front End systems.

initially been added.

Use full state names rather than abbreviated form in reporting possible values
bug 1413

Conflicts:
	src/slurmctld/job_scheduler.c
	src/slurmd/slurmd/req.c

Bug 1397

Bug 1384

Improve logging, comments, and formatting

This change prevents a new debug() message from being generated
if an explicit timeout was set, as used by END_DEBUG3.

Log long processing times with the verbose() call only if they
  take over 3 seconds (previously 1 second).
Log anything over 1 second using debug().
bug 1082

installing

If some tasks of a job array are runnable and the meta-job array
record is not runable (e.g. held), the old logic could start a
runable task then try to start the non-runable meta-job, discover
it can not run, and set its reason to "BadConstraints".

Test case:
Make it so no jobs can start (partition stopped, slurmd down, etc.)
submit a job array
hold the job array
release the first two tasks of the job array
Make it so jobs can start

This backports some logic already added to v15.08

Conflicts:
	src/squeue/print.c

Squeue modified to not merge tasks of a job array if their wait reasons
differ.
bug 1388