Improve watchdog reliability. Add panic button driver. Simplify some code repetition

[cmorse]

This should improve how reliable the watchdog checking is. I've had many times where the websocket would be disconnected, but the watchdog would not catch it. I updated the watchdog so that it is based on the last time that a msg was received from the websocket, rather than the last time that 'refresh()` was called

I've been using this code for 3-4 months and the new watchdog code has reliably recovered the websocket connection many times.

[samyun]

@cmorse planning on checking this in?

[cmorse]

I don't have write access to this repo

[jedbro]

@cmorse - Do we know who has write access?
I'd like to get this checked in and add support for additional new Ring devices.

[samyun]

@ardichoke has write access, looking at previous PRs in this project.

@csteele-PD and @djdizzyd are both listed as org members of HubitatCommunity with recent activity - either of you have write access and can look at this and #18?

[Banner-Keith]

Sorry guys. I didn't get any notifications whatsoever about this. I just saw the tag tonight in the community thread. This looks good to me.