rs.cchalf could be more useful if it looked for intensities

[kmdalton]

A user has requested to calculate CChalf from scaled, precognition output which is in the form of unmerged intensities. Applying French Wilson scaling may distort the results, so I'd like to modify rs.cchalf to look for intensities in the input files.