Update the NWB packaging to include the auto-water annotation

[arjunsridhar12345]

See this for more details: https://github.com/AllenNeuralDynamics/DF-Refactoring/issues/132#issue-5075609390

[arjunsridhar12345]

posting notes from discussion with @micahwoodard today 08/07. We should try use the response time from the response software event instead of using the time from the trial outcome which is usually towards the end of the trial. This could be help for better classifying the reward as auto instead of earned

[arjunsridhar12345]

reverting the above and will just rename the label to auto to keep this scoped

[arjunsridhar12345]

closed by #66