Added session level metric file for session dataframe creation

[nkeesey]

Added session_metrics for session level DataFrame creation with functions from old process_nwbs.py.

• added basic metadata features from process_nwbs/compute_df_session_meta as a placeholder
• added majority of performance metrics from process_nwbs/compute_df_session_performance that I found relevant in my analysis
• Some metrics may be out-dated, need review on which to include from original script/which to ignore
• Only added one new metric: probability_chosen which is the average difference between the chosen probability and non-chosen probability divided by the difference between the largest probability and smallest probability (for normalization between different stage difficulties). Hopefully this provides a better metric of evaluation independent of task/stage difficulty. See eq: = avg(p_chosen - p_non_chosen) / (p_larger - p_smaller).
• NOTE: Alex and I's discussed response metric is the same as the 'finished_rate_with_autowater'

Session level metrics:

Basic Metadata:

• Session start time, subject ID, session date, protocol, etc.

Trial Counts and Rates:

• Total trials, finished trials, finished rate (both with and without autowater).

• Reward trials, reward rate, foraging efficiency, and foraging performance.

• Reaction time (mean and median), early lick rate, and lick consistency.

• Rates of double dipping (lick switches) during finished, reward, and non-reward trials.

Future work:

• Cleaning up and merging code from process_nwbs to only update relevant metrics
• Further probability chosen calculations

DEMO
session_df = session_metrics(nwb)