auto water reward fraction

[micahwoodard]

In the current df implementation of autowater, the reward given is a fraction of the reward volume. In our implementation, this cannot be achieved because reward volume is at the task logic level.