Blender Git Loki

Cycles: Avoid using lookup table for Beckmann slopes on GPU

This patch is based on some work done in D788 and re-formulation from Beckmann
implementation in OpenShadingLanguage.

Skipping texture lookup helps a lot on GPUs where it's more expensive to access
texture memory than to do some extra calculation in threads.

CPU code still uses lookup-table based approach since this seems to be still
faster (at least on computers i've got access to).

This change gives about 2% speedup on BMW scene with GTX560TI.