Also some small kernel speedups elsewhere.

Also address some other issues that came up in code review.

Remove the beta parameter since it's always 1 anyway.

a dependecy on scikits.cuda and pycuda.

Update docs to reflect batches and add some fallback code to add batches of 1 to non-batched version.

Some improvements to the gradient, but it's still transposed and partially wrong.

Fixed "cublasHandle_t handle" declaration in cuda_ndarray.cuh.

Required for cuda_convnet compilation in pylearn2 in Windows