I can do that, but we first need to define this precisely.
In LMA* or pathA* (with a heuristic that is pathdependent), do we count the
first value for each state? the last value?
Should we count each state only once, or should we count each value?
Also, unless there's some trick I'm missing, this means we would need to keep a
vector of fvalues, each with a count of the number of states generated with
that value. We could merge the fvalue that were proven closed, but there might
be several open fvalues.
