Improvements for 1-intro-to-linear-algebra.ipynb

[vasilygrz]

Hi @jonkrohn,

Following the issue (https://github.com/jonkrohn/ML-foundations/issues/16) that I opened yesterday, I've made these changes:

1. Removed the notebook metadata with nbstripout in VS Code so you can clearly see them. If you know about a better method, please tell me because I tried nbdime without success (1,500+ insertions/deletions for adding a single word).
2. Added the text inside the code cells as comments, which contain:
   • Additional explanations, such as the fact that np.dot and np.matmul produce the same results for 1D and 2D arrays, but they behave differently in higher dimensions (L2531).
   • Alternative functions, like tf.tensordot for computing the dot product directly in TensorFlow, and solutions (L2984), where I showed how to multiply two matrices using the torch.einsum function.