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

Not yet implemented:

- Vectorization of quadrature loop
- Anisotropic amount of quadrature points per direction

René Heß authored 6 years ago and Dominic Kempf committed 6 years ago

Rename the function and remove the backend switch. Reasoning:

- This function was only called once without the backend switching
  mechanism.
- The function takes different arguments than the pdelab
  pymbolic_quadrature_position.
- It returns only one component of the quadrature point.

Rename the function and remove the backend switch. Reasoning:

- This function was only called once without the backend switching
  mechanism.
- The function takes different arguments than the pdelab
  pymbolic_quadrature_position.
- It returns only one component of the quadrature point.

They are only to a lesser extent important for our HPC applications
and the vectorization algorithm struggles with the complexity of the
search space.

Avoid explicit doubles.

Avoid explicit doubles.

They are not used in a senseful way anyway and loopy
stumbled over them.

This is currently only used for level 4 instrumentation,
as these start and stop markers were messing heavily with
dependencies in the sum factorization code. This way, the necessary but
ugly work of adding instrumentation is separated much more
nicely.

I was tired of the amount of ifs that changed behaviour
depending on stage 1 or 3. This is a cleaner approach.