[[Imported from SVN: r7348]]

[[Imported from SVN: r7347]]

[[Imported from SVN: r7346]]

[[Imported from SVN: r7345]]

[[Imported from SVN: r7344]]

[[Imported from SVN: r7343]]

[[Imported from SVN: r7342]]

[[Imported from SVN: r7341]]

using the ideas of Patrizio Neff.

In particular, this code should handle real 3d continua, as well
as Neff-style Cosserat shells.

This initial commit only adds a lot of infrastructure.

[[Imported from SVN: r7339]]

[[Imported from SVN: r7338]]

[[Imported from SVN: r7314]]

[[Imported from SVN: r7313]]

[[Imported from SVN: r7312]]

[[Imported from SVN: r7311]]

[[Imported from SVN: r7310]]

[[Imported from SVN: r7309]]

[[Imported from SVN: r7308]]

[[Imported from SVN: r7307]]

[[Imported from SVN: r7306]]

[[Imported from SVN: r7305]]

[[Imported from SVN: r7304]]

[[Imported from SVN: r7303]]

[[Imported from SVN: r7302]]

[[Imported from SVN: r7301]]

[[Imported from SVN: r7300]]

[[Imported from SVN: r7286]]

don't compute and measure on the last grid.  Since we are doing uniform refinement that last grid is equal to the reference grid.  Hence computing there will lead to an error of zero and is a waste of time

[[Imported from SVN: r7233]]

instead of a singular value decomposition.  This makes computing
a Hesse matrix roughly 20% faster.

[[Imported from SVN: r7232]]

[[Imported from SVN: r7231]]

using a finite-difference approximation only for the derivative
of the gfe gradients wrt coefficients.

[[Imported from SVN: r7230]]

[[Imported from SVN: r7229]]

[[Imported from SVN: r7228]]

[[Imported from SVN: r7227]]

[[Imported from SVN: r7226]]

[[Imported from SVN: r7225]]

[[Imported from SVN: r7206]]

[[Imported from SVN: r7205]]

[[Imported from SVN: r7201]]

[[Imported from SVN: r7196]]

[[Imported from SVN: r7195]]