Overload elementary operations

[jwallwork23]

Closes #138.

Note that about 1,000 lines of changes are automatically generated from the ftorch.fypp changes. So this PR isn't as enormous as it looks.

[jatkinson1000]

This seems sensible, and impressive 😄
What is the argument for making torch_tensor_from_array() no longer return a value but instead operate internally?
We can chat over lunch.

[jwallwork23]

This seems sensible, and impressive 😄 What is the argument for making torch_tensor_from_array() no longer return a value but instead operate internally? We can chat over lunch.

Thanks!

If you overload the assignment operator then tensor = torch_tensor_from_array(...) will call torch_tensor_from_array with the provided arguments and then call torch_tensor_assign, copying that into tensor. If this and other constructor-type procedures go from being functions to subroutines then we avoid unnecessary data copies.

[jwallwork23]

As discussed in today's meeting, we should check what happens if an operator that hasn't been implemented is used. I switched from Q = 3 * a ** 3 - b ** 2 to Q = 3 * a ** b - b ** 2, i.e., take a tensor to the power of another tensor, rather than to the power of a scalar. I get the following compile error, which seems clear enough:

Scanning dependencies of target autograd
[ 91%] Building Fortran object test/examples/6_Autograd/CMakeFiles/autograd.dir/autograd.f90.o
/home/joe/software/FTorch/src/test/examples/6_Autograd/autograd.f90:54:14:

   54 |   Q = 3 * a ** b - b ** 2
      |              1
Error: Unexpected derived-type entities in binary intrinsic numeric operator ‘**’ at (1)

[jwallwork23]

Might as well pre/post-divide by scalar, too.

[jwallwork23]

Post-divide implemented in b804d54. Left pre-division (i.e., scalar divided by tensor) for now as I don't think it's something people use often and I need to check what that actually does on the Torch side.

[jwallwork23]

In f0ed978 I reworked the autograd test so that it tests multiply and divide without changing expected values.

[jwallwork23]

Merged in the Windows fix from #207.

[TomMelt]

Thnaks @jwallwork23 , this looks great. Just a single comment really and I think it should possibly be a separate issue anyway. lmk what you think. I am happy to merge as is.

(edit: obvs once you address the merge conflict)

[jatkinson1000]

Thanks @jwallwork23
This is a huge step, and really great work.

I have left a couple of queries/suggestions, but the code itself looks good.
I will approve, so feel free to look at my comments as you please and merge.