CASIM coupling to prognostic precipitation fraction used with CoMorph

@MichaelWhitall is already working on this.

Since Feb 12, 2026.

### Are there any linked Issues or Pull Requests?

_No response_

### Brief description

With CASIM, the sub-grid fraction for rain and graupel is currently set to the max cloud fraction in the column above the current height. CoMorph currently works with a prognostic field for the sub-grid rain / graupel fraction, which it sets to a suitably small value where most of the precip is being produced by a fractionally-small convective cloud area.

Single Column Model tests using CASIM reveal that CASIM's existing assumption yields an unrealistically-large rain fraction in typical parameterised deep convection situations. Diffuse detrained cirrus readily spreads across the entire grid-area at cloud-top, which results in CASIM setting the rain fraction to 1 throughout the troposphere, even though the area occupied by the deep convection producing the rain is < 5%.

Unlike in the Wilson-Ballard microphysics scheme, the rain fraction does not have a strong influence on the assumed hydrometeor size distribution in CASIM. However it does have important effects on rain / cloud / ice overlap and on canopy throughfall versus re-evaporation where the rain falls on the land surface. Therefore we should try and couple CASIM consistently with CoMorph's precip fraction prognostic instead of using the max cloud-fraction in the column above.

This will build upon some of the relevant changes already coded in this branch:
https://github.com/MetOffice/lfric_apps/commit/cc8671059e549b1f8da41404255feed6aeccfa81

lfric_apps branch:
[vn3.2_casim_precfrac](https://github.com/MetOffice/lfric_apps/compare/main...MichaelWhitall:lfric_apps:vn3.2_casim_precfrac)

Equivalent changes in the UM:
[vn14.2_casim_precfrac](https://github.com/MetOffice/um/compare/main...MichaelWhitall:um:vn14.2_casim_precfrac)

### Further details of the issue.

_No response_