implement various failure logic for the exposure command

from sps-ics slack channel:

I'm currently fixing a few bugs in the spsActor exposure command, and I've been thinking about the design. I’d like your opinion since I’m deep in the code right now.
Currently, the design is that if one of the detectors fails early in the exposure, the exposure fails immediately (e.g., CCDs are cleared, and the H4 ramp is stopped). By "early," I mean before the shutter opens. These failures include: * CCD wipe failed

• H4 didn’t reach reset
• H4 didn’t reach the first read
• One of the shutters won’t open

Once the shutters are open, not much can happen, and the detectors are always read anyway.
I implemented this approach because I thought the observer would want to be informed right away to fix the issue and hopefully restart the exposure.
However, another option would be to keep going, even in the case of these early failures. We could notify the observer using the new warning system, and it would be up to them to decide whether to finish the current exposure, fix the issue and restart, or just keep going.
The second option offers more flexibility and shifts the responsibility to the observer. But these are critical errors, and if the observer overlooks the warning, we could lose a 900s integration on one camera due to a small bug that could have been fixed in 10 seconds, so it’s not risk-free.
 
Implement both scenario and add it as config knob in pfs_instdata