fix(Rosenbrock): saturation adjustment on max(S_ice, S_liq)

[haakon-e]

What this changes

Refines the EndStateSaturationAdjustment limiter so the substep increment is held at saturation over the more-supersaturated phase, max(S_ice, S_liq), instead of over ice alone. Below freezing ice carries the larger supersaturation, so the criterion reduces to the previous ice-only limit and the cold deposition-instability fix is unchanged. Above freezing it limits on liquid, so warm cells condense to liquid saturation; an ice-only limit would instead stop vapor depletion before liquid saturation and hold the cell supersaturated over liquid (too little cloud).

Docs & test

Adds a "mixed-phase saturation criterion" section to docs/src/RosenbrockNumerics.md with a simplified 0-D model that exchanges vapor with cloud liquid and ice only, illustrating the Wegener–Bergeron–Findeisen transfer (why the two saturation thresholds matter), and a framework test covering the warm (liquid) and cold (ice) limits.

Scope and a known refinement

Limiting on max(S_ice, S_liq) is a single shared scalar on the whole increment, which is what keeps the limiter stable. It is correct wherever a single phase grows from vapor. When both phases are supersaturated below freezing (a vigorous mixed-phase updraft core), it limits on the ice floor and so condenses the co-present growing liquid past its own saturation in one step rather than transferring it gradually; a fully per-phase floor (limiting only the supersaturation-driven tendencies of each phase) is a documented candidate refinement. The distinction is inactive at fine substeps and does not affect the cold-instability fix.

[haakon-e]

This change is part of the following stack:

• feat(P3): Gauss-Legendre quadrature, integrand breakpoints -> GL(6) default #741
  • feat(Rosenbrock): unified RosenbrockAverage substepping framework + 2M+P3 exact-AD mode #726
    • feat(Rosenbrock): 1M donor / coupled / exact modes on the unified framework #731
      • feat(Rosenbrock): post-solve per-process verbose diagnostics #732
        • fix(Rosenbrock): saturation adjustment on max(S_ice, S_liq) #737 ◀
          • perf(BMT): recover the primal tendency from the Rosenbrock Jacobian pass #742
            • feat(BMT): analytic ManualJacobian for the 2M+P3 Rosenbrock substep #743
              • feat(BMT): coupled Tier-3 donor linearization for the 2M+P3 ManualJacobian #744

_{Change managed by git-spice.}