Added brier score for p-metrics in forecasting

[jagjeet-singh]

This PR adds brier score based metric to forecasting. The two new metrics are: brier-minFDE = (1.0-p)^2 + minFDE and brier-minADE = (1.0-p)^2 + minADE

Motivation for using brier score for assessing uncertainty:

• Brier score is a commonly used metric in probability calibration ("Brier, G. W. Verification of forecasts expressed in terms of probability. Monthly weather review, 1950.")
• The contribution is in [0.0, 1.0], thereby making sure probabilities are not penalized too high.
• No tunable parameters.
• Given that both minFDE and f(p) are in a similar range, the methods will be motivated to do well in both.

Top methods from eval ai leaderboard:

[johnwlambert]

Thanks for the PR, @jagjeet-singh. Would you mind adding a few sentences here explaining the motivation for the change, and links to any relative literature? Maybe those could go into the docstrings, as well?

Which scenarios do we have tested so far?