Release 2.0.15

Fixes issues with potentially double counting dynamic resources with multi-node jobs.

Release 2.0.14

Uses new ScaleLib - including azpbs remove_nodes --permanent (and azpbs join_nodes --include-permanent) to safely remove a node without concern it will be added once again.
Allow hostname changes after the node has joined - we safely remove it then add a new record for the new hostname
Added azpbs online/offline
Avoids race condition where a node could reuse a stale definition of the previous hostname.

Release 2.0.13

Fixes various 'offline' node issues as well as adds support for making managing non-autoscale clusters (SuspendTime=-1) easier.

Release 2.0.11

Avoid terminating state-unknown jobs

Restore nodes we have set offline for being down

Release 2.0.12

Avoid terminating state-unknown jobs

Restore nodes we have set offline for being down

Release 2.0.10

Add option to ignore hostnames or any onprem nodes

Release 2.0.9

Fix bug with assigned/available shared resource calculation

Release 2.0.8

Bump version to 2.0.8

Release 2.0.7

Revert "Use scalelib 0.2.6 and use the new “hostname_reported” metada…

Release 2.0.6

Bump version to 2.0.6