Add more complex seat-selection behaviour to Patient agents

[thomasGodfrey]

E.g. Selecting seats that are more distant from other agents (potentially based on human-behavior responses to disease prevalence - see 'Modelling the influence of human behaviour on the spread of infectious diseases: a review'

[szschaler]

Very good idea. It may be good to implement the general capability in the Repast engine but expose a way for specifying strategies (and possibly stacking them) in the modelling language. A good starting point might be to list some realistic strategies so we can analyse differences and similarities.