Neural Quantum States (NQS) for Quantum Chemistry

[knitterAQ]

This PR introduces a new Tangelo example, which is a PyTorch script for training neural quantum states (NQS) ansatze, using autoregressive Variational Monte Carlo, to find ground state energies of electronic Hamiltonians. Tangelo is used to create the second quantized Hamiltonians (mapped to qubits via Jordan--Wigner) and perform HF/CCSD/FCI baseline testing.

[alexfleury-sb]

I did not scrutinize every line of code since I am not familiar of NQS workflows, but I did look at the README file, docstrings consistency and high-level stuff.

I think the coding style here is high quality, even though some styling is not consistent with Tangelo. If it works, I am ok with merging.

Before doing so, I would like:

• No more __pycache__ folders tracking.
• A mention in the README of where Tangelo is useful in the code.

[ValentinS4t1qbit]

2 things:

1. Let's put a clear disclaimer about our intent with this code in the readme, after the intro paragraph. My understanding was that we do -not- intend to maintain or further develop this code. Is that correct ?

2. I'm not sure I understand why there is a gitignore or some files (what is the .dat binary ? Why the .bak ? Why are some files mentioned in the gitignore present ?). Following our intent for the previous point, let's make sure the adequate files are present.

If one of the goals is to allow people to reproduce the research, I'd make sure to attach the whole code blob to the article on ArXiV, with all the artifacts etc.