ESM1.5 use of STASH code 3173

[MartinDix]

See problem description at ACCESS-Community-Hub/ACCESS-Archiver#4

Is this the only instance?

Use a model version dependent override of the variable metadata to support ESM1.5 and CM2 (and newer developments)?

[blimlim]

I think there's a similar issue with STASH code 33001, ATM TRACER 1 AFTER TS. um2netcdf4 gives this a standard_name of mole_fraction_of_ozone_in_air which I suspect isn't accurate. E.g. the output for this variable looks like:

while the ozone ancillary ozone_1850_ESM1.anc I think is given just as a zonal mean, and looks like:

In this case, stashvar.py doesn't specify a standard_name and so it looks like um2netcdf reverts to the name provided by iris.