Fix reset behaviour (with history fixed)

[CloCkWeRX]

#1273 but with history a tiny bit cleaner.

For devices configured with IDs to be sent with a RESET command, fix the logic. When i made the RESET logic less aggressive, I made the code only send it if no DPIDs were detected. However some devices (like power plugs), will present a DPID of 1 (the main switch), but not the additional power sensor DPIDs. The previous logic would never then trigger a RESET which would expose those additional DPIDs. This logic fixes that issue.

@rospogrigio - as discussed in PR #1087. This code stands separate from that one. I've tested this week and all appears to be functioning correctly on my system.

@sibowler - I think your branch just wants 1x more git pull from upstream.

Recommend squash merging to clean things up a bit further

[sibowler]

@CloCkWeRX - When I try to sync via github, it says i can't and that I'm already up to date with upstream?

[sibowler]

@CloCkWeRX looking to get this going again - I've pulled the latest code from main and there were no clashes of changes. In #1273 the validate-hacs check is failing due to a 500 server error. Any thoughts here?

[CloCkWeRX]

Well, these must be busted in master

Error: R] [SERVICES] Service reload has no name and is not in the translations file
Error: R] [SERVICES] Service set_dp has no name and is not in the translations file
Error: R] [SERVICES] Service set_dp has a field device_id with no name and is not in the translations file
Error: R] [SERVICES] Service set_dp has a field dp with no name and is not in the translations file
Error: R] [SERVICES] Service set_dp has a field value with no name and is not in the translations file

[sibowler]

@CloCkWeRX - Is there anything I can do to assist?