More realistic mask wearing (not for V1, probably for V2)

[DAM-Philippon]

Masks are an important policy of our model, as the scientific community is still on debate about their usage.
We currently represent masks by having one probability for an Individual to wear a mask that is computed at each step. This probability is, currently, not depending on the epidemiological status of the Individual: the probability is the same for symptomatic, asymptomatic and even not-infected people. While I think the last two should still have the same values (i.e. there should not be a difference in terms of behaviours between an asymptomatic/presymptomatic and not infected individual).
I think symptomatic Individuals should have their own probability of wearing a mask (having symptoms tends to make you wear a mask more than someone not having symptoms, depending on the population of interest), which could possibly impacted by policies (if you are tested positive, for instance).

[DAM-Philippon]

Another important aspect is that restaurant/bar/karaoke should make people remove their mask (you can't eat with a mask on). This is something that we've seen in HK, were 97.5% of people wearing mask, with the increase in clusters related to restaurant/bar.