nesting

in this branch nesting is possible in telemac2d and telemac3d. this means that there are boundary conditions that very in space and time. the format that is chosen is the format of the time series files. This means that you can generate a time series file in a large model and use it as boundary condition in another model

there are three possible options:

-read depth averaged (2d) variables in a t2d model (new feature).

-read depth averaged (2d) variables in a t3d model (new feature)

-read depth varying (3d) variables in a t3d model (existing feature). however, this feature is modified to be compatible with the others. Especially now the node numbers were to apply the boundary condition data are specified in the IKLE array (which has a changed shape), rathre than the IPOBO array. I could not use the IPOBO array, becausde that is used for gretel in generating the binary boundary condition files

In this way results of 2d and 3d model can freely be combined.

other enhancements include:

-possibility to combine binary boundary conditions with boundaries in the ascii (this is demonstrated in the example), such that you can add an upstream discharge to a river model

-possibility to add extra tracer variables that are not in the binary boundary files, which use the ascii boundary conditions or the values form the .cas file)

there are couple of things still todo, for which I will make separate merge requests (after my holiday):

-add documentation for the manuals

-update the current python scripts for dealing with binary boundary conditions (because the file format has changed)

-add a python script that generates the coordfile given the two meshes, that generate the correct time series file

@chi-tuan.pham . Should I also built in acheck in the fortran code to detect the old 3d format, to give an error message in case it is used