Feature request in phosphor-pid-control

[Tanous, Ed]

> 
> I’m pretty sure Matt Barth added support for
> xyz.openbmc_project.State.Decorator.OperationalStatus
> to phosphor-hwmon already.  It certainly doesn’t cover something like
> mapper timeouts though (nor should it, IMO).
> 

Interesting.....  I had no idea.  That I think that solves half of the issue (operational versus not), but still seems to miss the distinction between unavailable and error.

With that said, it sounds like it might solve the issue in the context of this discussion, so that's pretty cool.

-Ed