Keep the gradient flowing

new get_blas_funcs in scipy.linalg

Today got merged some changes I made to function scipy.linalg.get_blas_funcs(). The main enhacement is that get_blas_funcs() now also accepts a single string as input parameter and a dtype, so that fetching the BLAS function for a specific type becomes more natural. For example, fetching the gemm routine for a single-precision complex number now looks like this: [cc lang="python"] gemm = scipy.linalg.get_blas_funcs('gemm', dtype=np.complex64) [/cc] compared to the clumsy old syntax: [cc lang="python"] X = np.empty(0, dtype=np.complex64) gemm, = scipy.linalg.get_blas_funcs(('gemm',), (X,)) [/cc]