changed comment to tell that we make a special case for scalar in all case into elemwise for the gpu.

added a test not actif to compare the elemwise speed of the current version again another that I do that enable the collapse of contiguous dimensions.

in test_nnet use the new verbose flag and allow to use only the cpu version and allow to don't pickle.

If the THEANO_GPU env var is set to a number, we will use this card number.

some small fix to make it work when the end don't have CUDA_ROOT, but that the other env variables are set correctly.