Add a superstructure for repeated experiments and distributional output

[davidsiska-vega]

We'd want to run certain experiments multiple times (e.g. 30 so LLN / CLT starts to kick in).

At this point we'd want to be able to:

• collect average of values of interest + stddev so we can plot confidence intervals
• plot distribution

Additionally we'd like to be able to vary one (or two) parameters systematically and use the averages above mentioned to produce

• boxplot chart (different values on x axis, boxplots along y-axis
• heatmap plots (when varying 2 params)