reappearing_cross_section with spreading<1

[RichardPfr]

If the spreading is smaller than 1 currently the reappearing_cross_section calculation returns an empty array: https://github.com/pyroll-project/pyroll-core/blob/93bfbaf21d057b63e7d5023491470a41ca0b49d6/pyroll/core/roll_pass/hookimpls/roll_pass.py#L190-L192

Though spreading smaller than 1 is a circumstance that can realistically happen (at high rolling velocities for example) so the area difference should still be included somewhere by either:

• including it in the current reappearing_cross_section - though this would be semantically incorrect, as its not actually a reappearing cross section then
• add an additional value like "reduced_cross_section" (name is still misleading but i cant think of a better one) that only returns an array if the circumstance desribed above is the case.

I would say the first is still the better option, just so that bloat is reduced (and it also dirctly works with existing code). @axtimhaus / @ChRen95 What do you think?

[axtimhaus]

Unfortunately, shapely is not able to represent something like negative areas. One possible solution would be to make a new reappearing_cross_section_area hook which is float and negative in this case. But this would be hacky and unsafe.

I vote for the second alternative, as then the semantics are clear. Naming is tough though, also the names of the existing hooks are somewhat cumbersome and hardly expressive. The name reduced is ambiguous with the current displaced.

I have no better idea for now. Will think about it.

[ChRen95]

Well I think displaced and reappearing should stay. Making it a float value would mean losing all neat properties of a shapely polygon so I'm also not a fan.