Move the working_cluster_rec reference out of the following functions:
slurm_send_recv_controller_msg(), slurm_send_recv_controller_rc_msg(),
slurm_open_controller_conn_spec() and slurm_send_only_controller_msg()
so that communcations to multiple clusters can be performed in parallel
(using threads)

If a cluster in our federation is down, it's "control_host" value
is "" and an attempt to communicate with it will abort.

Previous content has news about version 16.05 release. Updated
for version 17.02 release.

[min|max]_cpus if not already set when a job comes in.

Change slurm_load_jobs(), slurm_load_job_user() and slurm_load_job()
to all use the same underlying function, which was formerly just
used by slurm_load_jobs(). This function will get data from all
of the clusters in a federation, but in a serial fashion for now.

The clang error isn't entirely clear, but this should fix it.

the job may still point to the qos_id of the removed qos.

the job_table.  This would represent the removal of runaway jobs.

I also made it work so the versions would be easy to see.  I couldn't easily
make this into multiple commits or I would had.

Add squeue --local and --sibling options to modify filtering of jobs on
    federated clusters.

Modify slurm_load_jobs() function to load job information from all clusters
    in a federation.

Fix double free issue from commit 3851f619

Same issue from fixed in a different spot with b19637bd.

Bug 3617

This reverts commit 8343365f.

Fix double free issue from commit 3851f619

Same issue from fixed in a different spot with b19637bd.

Bug 3617

Rename some scontrol show job fields to match sview
(and keep them in alphabetic order in the sview output):
ViableSiblings changed to FedViableSiblings and
ActiveSiblings changed to FedActiveSiblings

Added FedActiveSiblings, FedOrigin and FedViableSiblings to job info

steps.

The wrong function was being called to release memory.
Bug introduced yesterday in commit c8bf6b5d

The xmalloc was based upon the wrong data structure

No changes to logic, just added some parenthesis and brackets

Reroute federated scancel <jobid>'s to the origin cluster.

The local cluster will cancel the job if federated job is running on the
cluster, otherwise it will route the request, back through the client,
to the origin cluster.

Set the cluster lock even when the cluster is the only viable cluster.
The cluster_lock is used to determine if the cluster is running the job
or not.

when seeing if the cluster is the only cluster in the viable list.

Adding to be able to route cancel (and other future) requeusts to the
origin cluster in a federation.

Federation Reconciliation

When a sibling establishes a connection a silbing that sibling will then
reconcile jobs with the other sibling.

when the job is being purged due a the origin job being "cleaned"
(e.g slurmctld -c).