Use unique names for individual resources downloaded via dataset page

[alycejenni]

Currently the config means that all datastore files are downloaded with the name resources.csv. Ideally we'd use original filenames, but this would be a quicker change that would at least mean a user downloading the entire dataset doesn't keep accidentally overwriting their files.

Would just need to change the default download config for dataset pages to use "separate_files": true.