Heterogeneity slider choices

[dushoff]

The first screen seems too limited, maybe allow a choice of beta_mean?

The second screen seems OK for guided exploration, although it might be good to allow a choice of number of runs

The third screen should probably be a bit more limited (certainly not rho); we could add a fourth screen (with everything) for extra exploration (after the lab).

[pearsonca]

I roughly like this split, but I'm thinking:

1. show effect of changing R0, with a tiny bit of heterogeneity + stochasticity - so add choice of beta_mean, otherwise same
2. show fixed R0, changing heterogeneity + pop size - keep same
3. focus on R0 near 1 - cut most options
4. new optional "wide open" - essentially just a play ground

I think we need to generally add some theoretical spillover-failure probability calculations (based on homogeneous population) + visual indicators / annotations.