Standardize output order and nomenclature

[jmason42]

For simulation functions, I think the output should be the more standard (independent variable, dependent variables) pattern. E.g. time, and then counts.

In the single-step case, this would be the time of reaction, and then the counts (as a 1D integer vector) at the end of the step.

For a run-until case, it should be the time-histories, and then the count-histories (as a 2D integer vector with time points on the major axis).

[jmason42]

Forgot to mention: as a pattern of usage, I think we should be doing

(time, counts) = <whatever function is called>

to standardize output variable names. I think it's a good trade-off between exactitude and generality.

[prismofeverything]

Agreed, I picked arbitrarily so if you have a reason for one or the other please feel free.

[jmason42]

Addressed by #22