PyTorch Integration with Tensor Comprehensions

This release add PyTorch integration with TC. Here is what the PyTorch-TC package provides:

• inputs and outputs to functions are torch.*Tensors
• Integration with PyTorch autograd: if you specify forward and backward functions, you get an autograd function that takes Variable as input and returns Variable as output.
• autotuner results can be cached to a file (for reuse)
• Various test for ML layers are provided in test_python/layers/
• To make it easy to use TC, we provide conda packages for it. We will provide instructions on how to install the conda package soon.
• Documentation is also made available.
• A tutorial is also provided to help get started quickly.