more convenient MCMC

[teddygroves]

I tried to quickly add documentation for MCMC sampling and realised it is currently pretty convoluted.

I would like to refactor so that the user can do something like this:

from functools import partial

from jax import numpy as jnp

from enzax.examples import methionine 
from enzax.mcmc import run_nuts
from enzax.statistical_modelling import (
	ObservationSet,
	prior_from_truth,
	enzax_default_log_density
)
obs = ObservationSet(
    conc={"met-L": 0.01},
    flux={"AHC1": 0.5},
    enzyme={"MS1": 1.6},
    conc_error_sd=0.1,
    flux_error_sd=0.2,
    enzyme_error_sd={"MS1": 0.15}
)
fixed_parameters = {“temperature”: jnp.array(310.0)}
prior_mean, prior_sd = prior_from_truth(methionine.parameters, sd=0.1)
posterior = partial(
	enzax_default_log_density,
	fixed_parameters=fixed_parameters,
	obs=obs,
	prior_mean=prior_mean,
	prior_sd=prior_sd,
)
samples, info = run_nuts(posterior)

I think that is basically the most minimal it could be