Time series correction and standardization

Fix common problems with time series, such as gaps and duplicated dates. The final product is a smooth and continuous time series.

Time series have frequent problems such as gaps in time (ex : skipping a few days) or duplicated dates.

This function aims to identify these problems, and correct them. More specifically :

1) Fill the gaps in time so that it becomes a continuous series with a constant timestep

2) Order the data chronologically

3) Check if a given date appears multiple time. If a date appears multiple times, but the corresponding data is the same, the data

will be kept (only once, obviously). Otherwise (i.e. same date, different data), both data will be considered missing

4) Convert a time series from a specific time step to another (ex : daily to hourly). Holes will be filled with NaNs.