the big change below is enabling the Badnell DR rates. a good fraction of the test suite results changed as a result. the next merge will update part of the auto test suite, but not half. with this rev by itself the test suite will certainly not pass cleanly, but there should be no fpes.
zero.cpp - turn on Badnell DR by default
mole_co_etc.c - CRITICAL BUG FIX - make sure this goes into Robin's newmole
delay including ice formation on grains until both molecule and ion solvers have had at least one pass. Early evaluation before other solver had worked caused catastropic ice formation.
ion_solver.cpp - CRITICAL BUG FIX - make sure this goes into Robin's newmole
delay including source terms from chemistry solver until both ion and chem solvers have been called twice - prevents very large & bogus source terms from molecular network from upsetting the approach to a solution
punch_do.cpp
punch out radius command introduced
minor rewrite of if logic to avoid block nesting limits in vs2005
parse_punch.cpp
punch outer radius option
punch_colden.cpp
do not check on element and ion, so now possible to get molecular column densities
prt_comment.cpp
BUGFIX - fpe occured for VERY high ionization conditions where H0 abundance underflowed - protect against division
heat_sum.cpp
BUGFIX - fpe occurred for VERY high ionization conditions when AtomicCollidDensity? underflowed to zero - protect against division
dynamics.cpp
turn off CO network when advection is active - a conservation error developes which it cannot converge away from
conv_base.cpp
move IonChlor? to before CO_drive call so that all elements within co mole network are evaluated before CO called
