Allow acceptance probabilities different from $|\psi(s)|^2$.
In the proposed PR, acceptance probabilities are $|\psi(s)|^\mu$.
Consequently, all expectation values require reweighting with probabilities $|\psi(s)|^{2-\mu}$.
In this PR, samples are expected to be returned together with such a probability distribution (In the case of $\mu=2$, this distribution is equivalent to the uniform distribution).
Therefore, the differentiation between cases where a probability distribution is provided alongside the samples and those where it is not is no longer necessary. These functions have been deleted.
Two tests were added to certify the correct working of the sampler: A time evolution in the tdvp_t.py test file and a sampling test in sampler_t.py.
Quantities that are evaluated using the mpi_wrapper.py are expected to be arrays of rank 3, to allow for unified treatment of all cases.