Idea: Implement vkm's own sequential rejection algorithm generically for Venture

[axch]

Benefit: A presumably more scalable (than naive global rejection sampling) generic inference algorithm that also guarantees samples from the exact posterior.

Reference: Mansinghka, Roy, Jonas, Tenenbaum, "Exact and Approximate Sampling by Systematic Stochastic Search", AISTATS 2009.

Presumably will not handle continuous latent variables; oh well.

Suggestion: don't bother with the adaptation, at least at first.