"Inverter Command Delay" side effect

[Tixover]

I had an issue with some transitional glitches reading the data from the inverter some time back and increasing the Inverter Command Delay helped - however it does seem to be causing another issue because the coordinator refresh consists (in my case) of 8 sequential reads. It seems as if the coordinator_timeout doesn't seem to take into account the command delay and it is fairly easy to set it so the total refresh time is always going to be longer than the timeout (I seem to get repeatable Timeouts after every write).?

[WillCodeForCats]

I will look at the timing, it can be easily factored in.

[WillCodeForCats]

Will be added in PR #1099