Be careful when decreasing w_min or increasing w_inf

[gustavdelius]

When a user changes w_min, this does not automatically update the initial values, so after decreasing w_min the initial density in the new egg size class will be zero, and after increasing w_inf there will be no largest individuals. This may not be what the user expects. For now I am making no suggestion of what to do about this. Let's watch what happens in practice. But probably at least a warning message would be appropriate.