[hdpowerview] Optimize initialization of shades

[jlaur]

Fixes #12071 - please see this issue for further details.

Signed-off-by: Jacob Laursen [email protected]

This pull request fixes a concurrency issue and optimizes the status transitions for shades to come online as fast as possible:

• A five second initial delay in requesting shade data has been removed. It probably existed because of next problem:
• Bridge could receive data before shade things were initialized. This would cause them to come ONLINE prematurely, thus go back to UNKNOWN when being initialized. After this there would be a one minute delay in coming ONLINE again after next data poll.

Example of transitions for a single shade after the changes:

2022-01-24 23:28:01.835 [INFO ] [ab.event.ThingStatusInfoChangedEvent] - Thing 'hdpowerview:shade:hub:blind11' changed from UNINITIALIZED (BRIDGE_UNINITIALIZED) to INITIALIZING
2022-01-24 23:28:01.847 [DEBUG] [rnal.handler.HDPowerViewShadeHandler] - Initializing shade handler for shade 36321
2022-01-24 23:28:01.873 [INFO ] [ab.event.ThingStatusInfoChangedEvent] - Thing 'hdpowerview:shade:hub:blind11' changed from INITIALIZING to UNKNOWN
2022-01-24 23:28:02.010 [DEBUG] [ternal.handler.HDPowerViewHubHandler] - Child handler initialized: hdpowerview:shade:hub:blind11
2022-01-24 23:28:02.026 [DEBUG] [ternal.handler.HDPowerViewHubHandler] - Updating shade '36321'
2022-01-24 23:28:02.029 [DEBUG] [rnal.handler.HDPowerViewShadeHandler] - Caching capabilities 7 for shade 36321
2022-01-24 23:28:02.206 [INFO ] [ab.event.ThingStatusInfoChangedEvent] - Thing 'hdpowerview:shade:hub:blind11' changed from UNKNOWN to ONLINE

Planned for next PR in this area:

• Implement handling of shade removal, i.e. deletion of a shade in the PowerView Hub.

[jlaur]

@fwolter - sorry to exploit you a little, but #12061 (comment) seems to suggest that these channel state changes is your ball game, so maybe you could also have a look at this one now that we're at it? :-)