[petlenz]

Closes #115. Part of the uvwmaterial port (#94); stacked on #136.

uvwmaterial's eigenstrain_curing_material_base and degree_of_cure_dependent_material become two wirings rather than classes:

• Chemical shrinkage ε_ch = β ξ I is thermal_eigenstrain with the degree of cure as its scalar source ("temperature_source": "cure", "alpha": beta, reference 0); the integrator's state is a history like a temperature source's. Nothing new to port; the test pins it.
• Cure-dependent phase fractions: a phase reads its fraction from any producer (fraction_source / fraction_property, e.g. "cure", "state" of a scalar_stepper or a weighted_sum output); its volume_fraction output follows that property every update, scalar_complement_weight produces the matrix fraction 1 − c, and mean_field_tangent reads both. Cure-dependent moduli already go through weighted_sum.

Tests (test_curing_eigenstrain, 3): the phase fraction follows its source and the Mori–Tanaka bulk modulus follows the analytic formula at each value; constrained shrinkage gives −3Kβξ; fraction_source from JSON.

No report yet (time-driven point, #126).

924/924 tests pass (GCC 14, reports on).

[petlenz]

Review findings (design, tmech expressions instead of component loops, modern C++, comments, tests, correctness). Severity: must = wrong result, hang, build break or unreachable feature; should = design, duplication, accuracy; nit = style.

curing fractions / chemical shrinkage — ready with one should

• [should] volume_fraction stays required with fraction_source (dummy in JSON, can disagree until first update) → default 0 / exactly one of the two.
• [nit] sourced fraction never range-checked; pre-existing direction loop.

[petlenz]

Rebased on the Eshelby-node refactor (#128/#131): the curing fixture and the JSON test wire an eshelby_sphere node to the matrix and the inclusion names it; fraction_source is merged into the new inclusion. The volume_fraction-beside-fraction_source finding is still open.

[petlenz]

Producer/consumer pass: the curing inclusion gives its fraction only through fraction_source: "cure::state" (no dummy volume_fraction any more), the shrinkage coefficient comes through alpha_source.