PredictNewAreaMap should predict to new areas

[goldingn]

At present it only crops the training raster.

This could be achieved by passing in a raster via an argument, or by making a bespoke one for each covariate module (e.g. PredictNewAreaMapBioclim etc.)

[AugustT]

A catch all might be

PredictNewArea(newArea = ras)

where ras can be a raster (user provided), or a call to a covariate module:

PredictNewArea(newArea = Bioclim(extent = c(89, 90, -12, -13)))

[goldingn]

I like that - it allows people to pass a Chain of covariate modules too. It adds an extra layer of complexity to workflows though and might need some fairly nasty code to capture the argument and load the correct module.

If we changed the name to PredictNewCovariate or something similarly generic, that would be helpful. I expect people will want to use this for e.g. projecting to future climates (layer names would have to stay the same).

We had a user on the workshop who wanted to use zoon in her work, but needed this functionality - so a priority module!

[AugustT]

This has links with #121 in terms of the code required for the solution