Add a ScanAbort policy to improve robustness of scan on non-critical data acquisition failure
This is a feature request from ID10.
The basic idea is to have a way to separate critical counters (data acquisition failure should abort the scan) from not so critical ones (e.g. temperatures) for which a sporadic failure is acceptable (the data would then be missing). This would improve the robustness of long scan running by night.
The root cause is obviously the equipment that fails to respond from time to time (such as the Wago) but since it might be unavoidable, a workaround would be nice to have.
cc: @cammarat