Matrix inversion at code generation time does only work to
a very limited extent (up to n=4). We can instead assemble
the tensor in C++ and invert it there (e.g. using Dune::FieldMatrix)

Calculate the evaluation and jacobians of geometry mappings of cells
using sum factorization. By avoiding all geometry mapping of intersections
we can avoid the face-twist problem and just make the grid consistens (not
yet done).

TODO:
- Make grid consistent
- Vectorization
- 3D

It makes the code generation workflow more complicated without
contributing anything substantial

The introduction of FunctionView turned out to be a major problem
with more complicated forms. The original idea was to preserver the
structure of the finite element in a way, that loops over components
of a mixed element are realized by actual loops (treating them with
free indices and such). However, this causes quite some nightmares and
was never implemented as generically as needed (I even doubt that is
possible).

However, there is another option, which is to unroll any such loops
on a symbolic level. While this may sound like a bad idea at first
there is some really positive aspects about it:
* ListTensor and ComponentTensor nodes collapse completely (and would
  otherwise have a big nightmare potential)
* Symbolic zeroes do not generate code - important in hyperbolic problems
  where the system matrices are quite sparse or for axiparallel grids,
  where geometric quantities have many zeroes.
* The compiler would unroll these small loops anyway.
* TSFC (and I guess also FFC) do it the same way.

Implementing this required me to redo the form splitting algorithm.
I rethought it and integrated it into the main ufl->loopy visitor.

Moves application of transformation from micro to reference element from visitor to PDELab interface

... not going to the cell first.

This removes:
- The now obsolete quadrature_loop_statement.
- Quadrature weights are now a globalarg and we do not create a
temporary anymore.