Normal vector estimation based on plane-probing approaches

[nyorem]

PR Description

This PR adds new classes for estimating normal vectors on digital surfaces. The estimators are based on plane-probing approaches (see here, here and there). It specifically adds the following classes:

• DigitalPlanePredicate: aims to represent a digital plane as a point predicate,
• DigitalSurfacePredicate: aims to represent a digital surface as a point predicate,
• PlaneProbingTetrahedronEstimator: first category of estimators based on the deformation of a tetrahedron,
• PlaneProbingParallelepipedEstimator: second category of estimators based on the deformation of a parallelepiped,
• Classes for representing the candidate sets were also added (see the PlaneProbing*Neighborhood classes).
• PlaneProbingDigitalSurfaceLocalEstimator: adapter of the previous two classes to estimate normal vectors on digital surfaces.

It also adds new examples that illustrates how to construct and use those classes.

This PR also adds another estimator that is based on maximal segment computation on slices of digital surfaces. The PR could maybe be in split in two but the plane-probing estimator on a digital surface internally uses this other estimator (see the documentation for more details). The documentation for this part is still a work in progress. The new classes are the following:

• ArithmeticalDSSComputerOnSurfels: version of ArithmeticalDSSComputer that works on projection of 3D surfels,
• MaximalSegmentSliceEstimation: the new estimator class.

Checklist

• Unit-test of your feature with Catch.
• Doxygen documentation of the code completed (classes, methods, types, members...)
• Documentation module page added or updated.
• New entry in the ChangeLog.md added.
• No warning raised in Debug cmake mode (otherwise, Travis C.I. will fail).
• All continuous integration tests pass (Travis & appveyor)

[JacquesOlivierLachaud]

Great work ! Very clean with a clear structure. My comments are mostly on the form.

[JacquesOlivierLachaud]

I let you closed the last conversations once you are done with your changes.
I will merge afterwards.

[nyorem]

All the changes were made and all checks passed!

[JacquesOlivierLachaud]

Everything is fine.

[dcoeurjo]

@nyorem @JacquesOlivierLachaud the testPlaneProbingParallelepipedEstimator is way too long for a unit test. Could you please reduce the size of the geometries you are testing?

on travis:

   233/233 Test #122: testPlaneProbingParallelepipedEstimator ....   Passed  727.51 sec