Change object/morphism types in model notebooks

[quffaro]

Closes #184.

[epatters]

Thanks Matt for taking this on!

I have one high-level comment about the architecture, reflected in several of the comments below. Rather than using the cell constructor mechanism to construct a whole new cell and then extract its object/morphism type, we should, in each object declaration cell, create a cell switcher by looking up the available object types directly from the theory, and similarly for the morphism declaration cells. No code in the notebook directory should make reference to theories or models.

[epatters]

I don't think it makes sense to implement this feature by changing modelCellConstructors. As the name suggests, "cell constructors" are functions that construct new notebook cells. This is quite different in scoping to mutating existing cells of a given type.

[epatters]

A hint that this design is off is that notebook editor is not supposed to know anything about theories, models, etc. It's a completely generic component.