Derived precipitation diagnostics

### What problem does your feature request solve?
precipitation coming straight out of the model needs to be postprocessed to give precipitation diagnostics.

- [ ] We want total_rainfall_rate consisting of large-scale and convective rainfall rate in a ppn section to compare GAL vs RAL. This requires adding convective and stratiform ppn for GAL to calculate total rain. RAL only requires surface_microphysical_rain (large-scalerain) as convective rainrate not switched on in RAL.
- [x] LFRic: surface_microphysical_rainfall_rate [kg m-2 s-1] = large-scale rainfall rate Ls_rain STASH 4203 (m01s04i203))
- [ ] LFRic: surface_convective_rainfall_rate = conv_rain (stash m01s05i205)
- [ ] LFRic surface_rainfall_rate = total_rain (STASH 5214 (m01s05i214))
- [ ] LFRIc precipitation_rate = total_prec (total_rain + total_snow, STASH m01s05i216) this includes rain+snow and LS+convective
- [x] convert units to mm h-1 rainfall per hour
- [x] set colourscheme for [0.125,0.25, 0.5, 1,2,4,8,16,32,64,128,256]
- [x] custom colour

THE RNS outputs LS_rain and LS_snow.

### Describe the solution you'd like
reuse existing recipes for quicklooks but hardcode total_rain, large-scale rain and convective rain ppn variables to give ppn diagnostic. Caveat if convection scheme switched off in regional runs then convective rain might not be output under this diagnostic but as part of ls_rain.
include new rose gui for precipitation related diagnostics insetad of piping through quicklook recipes

### Describe alternatives you've considered
considered piping through quicklook recipes, but don't intend to have logic in recipes, but instead in operators.

Translation table from LFRic to UM