[Rivet] Weighted profile errors fixed / long-overdue release

[Andy Buckley]

Hi everyone,

I just fixed the remaining user-visible problem that I'm aware of in
Rivet... typical, I spend forever working out the most general form of a
weighted standard error and then go and write the crucial term upside
down in the code! Anyway, profile errors in weighted events look sane now.

So, I think we're now ready for a very long-overdue release. Frank has
reverted my fully-justified and fully-destabilising re-definition of
shared_ptr comparisons, and while this masks an annoying memory
management issue it does mean that the current trunk is stable from a
usability and results correctness point of view. I'll roll up my sleeves
and fix the internal memory horrors before the next release (which will
*definitely* be YODA-time!)

So, other than adding proper underscore handling to make-plots (which
Hendrik has said he'll do, hint hint), I don't think there are any
issues remaining (at last) to block the release that we aimed to make in
June! Any opposition to me tarring it up and installing in Genser once
the make-plots underscore thing is sorted?

Andy

-- 
Dr Andy Buckley
SUPA Advanced Research Fellow
Particle Physics Experiment Group, University of Edinburgh

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.