bug fix in _freesurface.py and add examples

[NengLu]

1. fix a bug in _freesurface.py. TField should be set back to the initial mesh before advect the top nodes and remesh.
2. add 3 examples to docs/examples: (1) Kaus2010_Rayleigh-Taylor_Instability (2) Crameri2012Case1_Relaxation (3) Crameri2012Case2_Rising_Plume

Reference

1. Kaus, B. J., Mühlhaus, H., & May, D. A. (2010). A stabilization algorithm for geodynamic numerical simulations with a free surface. Physics of the Earth and Planetary Interiors, 181(1-2), 12-20.
2. Crameri, F., Schmeling, H., Golabek, G. J., Duretz, T., Orendt, R., Buiter, S. J. H., ... & Tackley, P. J. (2012). A comparison of numerical surface topography calculations in geodynamic modelling: an evaluation of the ‘sticky air’method. Geophysical Journal International, 189(1), 38-54.

Results of Kaus2010_Rayleigh-Taylor_Instability at Time = 5 Ma (dt = 2.5 kyr)	before	after

[rbeucher]

So copying the mesh did not work?

[NengLu]

So copying the mesh did not work? mesh.data.copy() didn't work. and there are no function of mesh.copy() or copy(mesh),

[rbeucher]

from copy import copy

copy(mesh)

[NengLu]

from copy import copy

copy(mesh)

Humm, weird, that still didn‘t work. Here are the mid of the RTI model mesh nodes coords, which should be around 0.500 while it deforms too much.

[rbeucher]

Try with

from copy import deepcopy

[NengLu]

Try with

from copy import deepcopy

deepcopy can't work on model.mesh object, shows error: TypeError: can't pickle SwigPyObject objects

[NengLu]

Can we create the init_mesh in the _model.py below the def freeSurface, make it only works when the users using the freeSurf=True. It seems an easy way to build when the copy(mesh) didn't work, and more clear than the changes in commit https://github.com/underworldcode/UWGeodynamics/commit/43374a2c0bee06987784f34a0af29b157d763672

    def freeSurface(self, value):
        if value:
            self._init_mesh = 
            self._freeSurface = FreeSurfaceProcessor(self)

[rbeucher]

No because it goes against encapsulation. I would rather have the code as it is now.

[NengLu]

No because it goes against encapsulation. I would rather have the code as it is now.

Okay, thanks.