Experimental design - expecting singular observation vs continuous

[SCYarnes]

DeltaBreed currently doesn't support accepting observations via BrAPI, in part because we lack a way to validate expected vs unexpected repeated observations. When we plan our experiments we would like to be able to specify expectations by observation variable, so that when we receive a repeated observation we can generate informative error messages to alert users when there are unexpected repeated measures due to errors like mislabeling or mis-orientation in the field.

We do not want to duplicate ontology variables to achieve this goal. A single observation variable should be able be specified as singular within the context of one trial/study and continuous in others.