Allow `population` to be a list-like in epidemic models

[pratikunterwegs]

This issue requests that epidemic models should accept lists of <population>s to quickly simulate uncertainty or different assumptions regarding population characteristics, such as the demographic stratification of social contacts, or initial conditions.

[pratikunterwegs]

Focusing on single country modelling so this issue is not currently a priority.