issues around physical interaction data ftp and download

[ValWood]

We should make it clearer how the HCPIN network data was created and why, by:

i) linking from the documentation to the biological slim page where the networks are available
ii) explaining the HCPIN dataset is a stop gap between networks and pathway models
iii) Improve the documentation surrounding interaction data more generally to explain the difference between the bioGrid export and the HCPIN data content (perhaps a page for "interaction data")
iii) include the GO evidence codes in the HCPIN datasets if we keep them public (BUT the files for export to esyN mustn't be edited as it will break, and there is nobody maintaining this)