WFAF-D: num generations (G) value of 1 should be allowed

[jasondk]

• This seems to be flagged as an error but should be allowed
• Actually, I think a value of 0 should probably be allowed as well (for computing jump probabilities from one model to another - we can discuss if this is unclear): use case might be to start in initial probability distribution given N=1000, then jump to N=100 in 1 generation and output probability distribution after 0 generations in N=100