MIMICS overflow respiration not being handled correctly

[wwieder]

Brief summary of bug

MIMICS overflow respiration isn't being called, allowing microbial (and then soil) C:N ratios to get all out of wack.

I'd also suspect this is what's making preliminary MIMICS runs so nitrogen limited (a feature that has been stubbornly insensitive to parameter changes). Thanks to @katierocci for helping identify the problem and @slevis-lmwg for helping with a potential solution

General bug information

CTSM version you are using:
ctsm5.2.028, which works with Anderson acceleration, but present in all MIMICS code since it was introduced.

Does this bug cause significantly incorrect results in the model's science?
Yes,

Configurations affected:
Any using MIMICS soil biogeochemistry namelist option

Details of bug

In SoilBiogeochemPotentialMod we're turning off the immobilization flux in MIMICS by setting p_decomp_cn_diff_ratio and then pmnf_decomp_cascade <= 0`. This is fine, and consistent with assumptions in MIMICS that microbes get first crack at the soil N they are decomposing.

BUT if inputs to microbial biomass > microbial stoichiometric demand then immobilization should occur and this ratio is > 0. Under these conditions 'overflow' respiration should occur, allowing microbes to blow off the extra C they don't need.

Overflow respiration happens in BiogeochemCompetitionMod but only if (pmnf_decomp_cascade(c,j,k) > 0.0_r, which is therefore never triggered.

Finally, if overflow respiration was being set correctly, it would modify HR C fluxes and inputs to receiver (microbial biomass) pools correctly in SoilBiogeochemDecompMod.

I'll open up a PR before long to address the bug.

[wwieder]

@mvdebolskiy, can you let Betty know about this (I'm not sure of her github hanle)

[mvdebolskiy]

@elisacw FYI.