Add P3 as the microphysics scheme

[crterai]

This PR adds P3 as the default microphysics scheme used for the v3atm. P3 is a more advanced cloud microphysics scheme in ice microphysics: remove the artificial conversion between ice categories and a few other tunable parameters (related to ice particle fall speeds and density) and account for riming particles.

! This will change the default microphysics from MG2 to P3. Therefore, it is a non BFB change to the code.

More details of the scheme can be found here: https://acme-climate.atlassian.net/wiki/spaces/NGDAP/pages/3518496896/P3+integration

*Completed

• Port P3 code on top of NGD_v3atm branch.
• P3 is made to be default microphysics for NGD branches.
• Minimal (custom-10, custom-30) tests passed running with P3 using F20TR_chemUCI_Linozv3 compset

*Not completed yet

• Connections between aerosol scheme and P3. <- unclear if this is necessary or not

[Climate Changing]

[tangq]

@crterai , the changes under chemistry/mozart seem not related to gas chemistry. Do you see conflicts in these two files (chemistry.F90 and mo_gas_phase_chemdr.F90)?

[crterai]

@tangq - The changes in those files must be related to the aqueous chemistry fix that Mingxuan had mentioned in an email thread. And no, there were no conflicts that I needed to resolve in those files.

[crterai]

@wlin7, I believe this PR is ready to be merged. If you also think so, please merge it to the NGD_v3atm branch.

[wlin7]

Ok, @crterai . I will do shortly. Baseline tests (comparison) after yesterday's PR 15# still have some in the queue.

[wlin7]

Merged to NGD_v3atm.

[wlin7]

Note that the following threading test failed after this PR. The test passed before the merge. This is an known additional threading issue with the P3_PR.

PET_Ln5_P512x2.ne30pg2_EC30to60E2r2.F20TR_chemUCI-Linozv3.cori-knl_intel.eam-20tr_v3atm