Emit warning when branch limits equal 0

[jeanpaulwatson]

Egret mirrors Matpower convention, which treats 0 as unlimited in the case of branch limits. This is a mildly weird convention, and I as a user would like to know when a case has such data. Thus, I suggest emitting a warning for such branches, or at least adding a flag somewhere in the logic to force suppression - thus requiring a user to knowingly allow the convention to be applied.