so that the ordering code at G-code export may work directly with
pointers to PrintInstances instead of with pair of <PrintObject, copy idx>.
Also the PrintInstance knows its source ModelInstance, which allows
sorting of PrintInstances for sequential printing in the order
they appear in Plater's object list.
Refactoring of GCode export of color changes, extruder switches etc,
so that the "color change" like extruder switches are applied first
at the Wipe Tower / G-code export, so that adding / removing
an extruder switch at the G-code preview slider does not invalidate
slicing.
1) Don't allocate ExtruderOverrides if not necessary
2) Use boost::container::small_vector<int32, 3) for ExtruderOverrides
(usually less than 4 instances are printed))
Fixed layouts after the second opening of the dialog.
(Removing any extruder from the sequence does not cause an incorrect layout)
Validation of entered values added
(0 is not a valid value)
- abstract class WipeTower and its descendant WipeTowerPrusaMM were merged into a single (non-abstract) WipeTower class
- all uses of WipeTower::xy struct were replaced by Eigen Vec2f (it is no longer necessary to be independent on libraries that PrusaSlicer uses)
- the WipeTowerPrusaMM.hpp/.cpp will be renamed in the next commit (hopefully it will retain its git history that way)
The priming extrusions were handled separately from the rest of the wipe tower toolchanges. In order to be able to use the logic from previous commit for them (custom toolchange gcodes etc), some unpleasant code shuffling was needed
This is a fix of a long standing bug, where an extrusion is incorrectly
drawn from the end of the last wipe move. Interestingly enough,
this bug is in Slic3r at least since 1.2.9, but lucky enough
it only occured for single perimeter, no infill prints with wipe after
retract enabled, and only if the two successive slices were discretized
exactly the same, which is quite unlikely.
