ENH: better document data for default wflow setup

[DirkEilander]

Some datasets like LAI have undergone some preprocessing to derive monthly climatology, see #156
Other datasets like MERIT_Hydro have a few derivative layers (basins, stream order, etc) that are not well described, this relates to #89 .
It would be good to better document or even script the steps taken before this data can be used in hydromt-wflow.

Known issues:

• MODIS LAI
• MERIT Hydro basins index
• MERIT Hydro additionnal layers (stream order, basins)
• MERIT Hydro vrt format (combining tiles yourself is often prone to error)
• era5 orography layer
• reservoirs / lakes database are processed from GRanD and Hydrolakes

[DirkEilander]

A first step could be to add a table in the user guide with the default data sources and which setup methods these are used for. We can additionally share some data that is not currently available for direct download (e.g. MODIS LAI) via Zenodo. For MERIT Hydro we should link the 30 arcsec IHU version which includes a basin index file.

@JoostBuitink @hboisgon Let me know what you think?

Related issues #178

[hboisgon]

@DirkEilander indeed good to document better where the default input data for wflow is coming from. The table you mention is a good idea but I think some things are also good to mention directly in hydromt core repository (eg update link for basin index or era5 orography source). Also if more users than wflow use these datasets then they'll have the same questions.

So good to also find a way to document this better in core (at least fixing the broken link), or maybe just allow for hydromt to do some of the processing steps ?
For example, lakes is mainly about filtering column based on attribute see hydromt:#97 and maybe we could create the monthly average LAI maps in setup_laimaps method and then reference the original dataset as well in the data catalog ?