Let the TheaterCommander manage front line stance.

This improves the AI behavior by choosing the stances non-randomly:

* Breakthrough will be used if the base is expected to be capturable and
  the coalition outnumbers the enemy by 20%.
* Elimination will be used if the coalition has at least as many units
  as the enemy.
* Defensive will be used if the coalition has at least half as many
  units as the enemy.
* Retreat will be used if the coalition is significantly outnumbers.

This also exposes the option to the player.

game/commander/tasks/compound/defendbase.py

@@ -0,0 +1,19 @@
+from collections import Iterator
+from dataclasses import dataclass
+
+from game.commander.tasks.primitive.cas import PlanCas
+from game.commander.tasks.primitive.defensivestance import DefensiveStance
+from game.commander.tasks.primitive.retreatstance import RetreatStance
+from game.commander.theaterstate import TheaterState
+from game.htn import CompoundTask, Method
+from game.theater import FrontLine
+
+
+@dataclass(frozen=True)
+class DefendBase(CompoundTask[TheaterState]):
+    front_line: FrontLine
+
+    def each_valid_method(self, state: TheaterState) -> Iterator[Method[TheaterState]]:
+        yield [DefensiveStance(self.front_line, state.player)]
+        yield [RetreatStance(self.front_line, state.player)]
+        yield [PlanCas(self.front_line)]