burst_buffer/cray: Support file staging when job lacks job-specific buffer
    (i.e. only persistent burst buffers).
bug 2113

No change to document contents, just reorded the fields documented.

bug 2143

Always log some operations (previously logging was only triggered
  by long run time of the operations)
Log some operations with info() rather than debug() for DebugFlags=
  burstbuffer

test more association/qos limits before reserving resources for a
  pending job. This complements commit dcc943b7
bug 2129

Backfill scheduler: Test association and QOS node limits before reserving
    resources for pending job.
bug 2129

Bug 1493

Prevents the following sequence from causing a segfault:
$ scontrol create partitionname=stuff nodes=ALL
$ sbatch --wrap="hostname" -o/dev/null -p stuff
Submitted batch job 1047468
$ scontrol delete partitionname=stuff
$ scontrol update jobid=1047468 partition=stuff

tree.

Previously only SlurmUser could do so.

Bug 2006

Bug 2006

cgroup system for the extern step.  This needs to happen for accounting
purposes, but also makes things simplier.

step.

Test if job is pending before looking at burst buffer specs when
building the job queue. This should result in a slight speedup in
the job scheduling logic.

Previously only supported by SlurmUser and root.

Previously only reserved space for one task of pending job array.

Conflicts:
	NEWS

Support taking node out of FUTURE state with "scontrol reconfig" command.
Previous logic would keep node in FUTURE state if that was the original
configuration when slurmctld started. If job was running on the node,
it will stay running, but the node make not be visible.

Make SLURM_ARRAY_TASK_MIN, SLURM_ARRAY_TASK_MAX, and SLURM_ARRAY_TASK_STEP
    environment variables available to PrologSlurmctld and EpilogSlurmctld.

We needed to send a finish from each node in the step whether it had
any activity or not.  This way the controller knew things were done
on the node and the data was sent to the database.

Bug 2097