Problem 15 1D Convolution: Fix convolution inconsistencies

[spatters]

There are two problems with the convolution definition in problem 15

1. j index should run from 0 to K-1. I have corrected this.
2. The function f calls np.convolve which flips y before applying the convolution. So the function behavior is inconsistent with indexing in the convolution definition. Can either change the definition or change f to use np.correlate which does not flip. I have done the latter as I think not flipping is the standard in deep learning.