More Anisotropic Peeps

I've been messing with anisotropy some more and I finally got image based anisotropic lighting going. Previously it was incorrect (and dependent on a non-existent light vector) but now it's all awesome. Even the normal-mapped version works properly (I think, I can't really tell honestly). Next up I'm going to try straight up blurring of the environment sampling perpendicular to the groove directions to test quality and speed differences.