Lack of implicit unit conversion for attributes

[henrikt-ma]

The unitful literals arrived with implicit unit conversion for binding equations:

When the modified component of a binding equation has a unit-attribute, a binding equation consisting of a unitful literal or a negated unitful literal is implicitly converted to the component’s unit.

Given this, it is surprising that there is no analogous implicit unit conversion for the { start, min, max, nominal } attributes.

[HansOlsson]

I agree that it is a simple omission that should be corrected.