New HDD/CDD calculation added

[hagento]

A new approach for calculating HDD/CDD's has been added - calcHDDCDD, calcBAITpars - that bases upon the bias-adjusted internal temperature (BAIT). The calculation relies on processing raster data from the ISIMIP repository which are defined in the inst/extdata/sectoral/filemappings directory. These are read in using the functions readISIMIPbuildings and convertISIMIPbuildings which are tailored for pre-processing the raster data optimally for the above named purposes.
The code is working but partially throws memory conflicts that still need to be resolved with further development.

[robinhasse]

Maybe avoid this kind of function name to avoid confusion with madrat calc functions?

[robinhasse]

Are you sure you that have another function called time? If not, drop the package name.

Suggested change

	terra::time(hddcdd) <- as.Date(dates)
	time(hddcdd) <- as.Date(dates)

[robinhasse]

Very impressive work so far, @hagento ! I mainly looked at calcHDDCDD so far. Here, just a couple of general remarks:

• You did a good job already in breaking down calcHDDCDD into smaller subfunctions. I feel readability would improve a lot, if you go a bit further: Move all subfunctions that only work with arguments directly passed to them (should be most) outside of calcHDDCDD but still in the same file. The I would give all of them individual rogygen header where you specify all parameters and the output. That would help a lot in understanding the code.
• Try to avoid function calls with explicit package prefix like package::function() whenever possible. If you have no name conflict, move them to the roxygen header.
• I would recommend running linter on your files from time to time (in RStudio: Dropdown on Addins -> Lint current file). You have some coding style issues, but that is not as important now. But you are no allowed to use certain apply functions which I believe you currently do. Rewriting this with allowed function can be annoying so it is good to save the time.
• There seem to be some committed mappings that should rather go to the EDGE-B input integration PR

[robinhasse]

Great update, @hagento ! Documentation is so much better now. Please rename all functions starting with calc to something else (eg get) to avoid confusion with madrat functions - apart from calcHDDCDD obviously. This is not just about users reading the code but also madrat's logic. madrat::getCalculations would falsely list the functions for instance.