Python part still needs to be fixed, almost there, compiles in clang again

Thought it would maybe help with compilation

But not clear why this changes anything...

Fix missing inline

Added multipolar terms; doesn't compile on mac anymore

Appears to be working, additional testing required. Generalizing to handle G&V and G&T correlation integrals remains

Very subtle problem with handling derivatives at zero density

Closes #38

Closes #40

Helpful for debugging

Tests should pass again

Still can't trace isotherms with mixtures yet

Ready for Python!

Using sigma in A seems to be necessary to avoid underflow in some numerical calculations, as is done in PC-SAFT

Still need the model and so on, but working