Update to EUDEM slope data in pre-processing

[brynpickering]

This is the first step in the workflow to update to using EUDEM slope data, which a user would need to manually download from Copernicus. Full European coverage is 1.8GB, which seems quite manageable. This does away with dependence on SRTM/GMTED elevation data and accordingly a lot of processing steps. It also allows Iceland and upper Nordics to be included at full resolution.

This is a draft PR as the downstream handling of this data is not yet clear to me (as I detail in #4); I would appreciate guidance on this @timtroendle.

[timtroendle]

See my latest comment in #4. This PR should only change the data source in my opinion, without any downstream changes necessary.

[brynpickering]

Although this now is functionally only changing the data source, it also provides the option for more complex slope-based processing in future. However, the initial threshold rule (tech_slope_thresholds) is slow (~20 mins per job). Although it is parallelised, it involves a lot of read-writes, which have a time penalty on the cluster. My hope is that there is a way to parallelise only the vector operation, and then have only one read-write per rule. However, since we don't need the functionality provided here as of yet, I am going to leave this PR as-is, and suggest we consider #24 in its stead.