The spam is already being processed for links. There will only need to be a couple of extra steps to be able to report the actual spamvertised site instead of just the redirection service.
1. Match the link domain against a list of known shortening domains
2. Optionally, opportunistically check an unknown domain to see if it is a URL shortening service if the domain part is <= 9 characters or so — this will help so that nobody has to manually add domains to the list, but it may add too much overhead or garbage in the lookup table, slowing things down
3. Do a HEAD request and cache the resulting Location header
Unlike the problem of rotating DNS records, shortening service URLs are idempotent, so they really only need to be looked up once. Obviously, of course, this proposed method won’t catch ANY redirect, but it ought to take care of most of the shortening services.
I know that SpamCop does not consider the spamvertised site reporting to be a particularly important or high-priority thing, but I would love to see this nevertheless if attitudes ever changed in this regard.
