Prescale and hlt1_filter for two high-bandwidth QEE full lines

FYI @mvesteri @wbarter @lugrazet @cagapopo (sorry for another MR to your list)

Another prescale for QEE: this'll save 800 MB/s to tape when we go to high mu, and probs O(20) MB/s to disk (we already prescaled this to disk once).

Second commit: add a hlt1_filter_code on an expensive NoMuID line. Expected savings here are less clear but we already do this at the analysis level - it should make a decent difference and there's no reason not to push it up into the trigger.