Improve link_to_face_graph API

[sloriot]

also use it in Tetra remeshing to illustrate how to detect the sharp features of the domain

[sloriot]

@janetournois The output (even before my change) does not look correct. Would you mind having a look and let me know if it is as expected. My issue is that target_edge_length does not seem to be taken into account on the domain boundary.

[sloriot]

Successfully tested in CGAL-6.1-Ic-189

[janetournois]

Indeed there is a problem with the surface of the cube, but it is also in master. So it's not blocking this PR, and I'm taking care of it in another branch. Thanks @sloriot for the report

[janetournois]

Indeed there is a problem with the surface of the cube, but it is also in master. So it's not blocking this PR, and I'm taking care of it in another branch. Thanks @sloriot for the report

There is actually no problem here. The issue with the example is that the density of the input is very different between the surface and the inside of the cube. One iteration is not enough to uniformize edge lengths. Also note that the target edge length (0.02) is very small compared to input edge lengths (even inside) so you need quite a few iterations to reach it.

I will push an improvement to this PR

[github-actions[bot]]

This pull-request was previously marked with the label Tested, but has been modified with new commits. That label has been removed.

[sloriot]

Successfully tested in CGAL-6.1-Ic-208

[sloriot]

For some reason, it seems to be responsible for time out in Tetremesh testsuite.