Support Custom Statistics For Python

[AlexCatarino]

Expected Behavior

Use case: we don't want to consider non-trading and/or no-holdings days in the Statistics.

Actual Behavior

Only C#.

Potential Solution

Create IStatisticsServicePythonWrapper and add PyObject overload to SetStatisticsService.

Reproducing the Problem

New feature.

Checklist

• [x] I have completely filled out this template
• [x] I have confirmed that this issue exists on the current master branch
• [x] I have confirmed that this is not a duplicate issue by searching issues
• [x] I have provided detailed steps to reproduce the issue

[jsehnert101]

Requested custom statistics in Python to better reflect the underlying characteristics of a trading strategy. For equity-based trading strategies, there is no/negligible opportunity cost during non-trading days (i.e. weekends), so their inclusion will underestimate volatility and average returns, consequently impacting metrics like the Sharpe ratio. In the presence of multiple return streams, one may be less concerned with the opportunity cost of their investment / cash positions of a single return stream - particularly those that are discreet in nature - and more concerned with capturing its behavior in absence of an opportunity cost. Custom statistics will allow users to make their own assumptions on the meaningfulness of the aforementioned concepts.

[Martin-Molinero]

This use case will be covered by https://github.com/QuantConnect/Lean/pull/7690, where we are adding a new algorithm setting TradingDaysPerYear where the user can overwrite the default value used for calculating statistics, default to 252, except for crypto brokerages which will use 365