[petlenz]

Closes #111. Part of the uvwmaterial port (#94); stacked on #133.

materials/incremental_linear_elasticity.h (incremental_linear_elasticity): σₙ₊₁ = σₙ + C : (εₙ₊₁ − εₙ) with stress as a history property and the tangent read from tangent_source (any material providing one), so a tangent that changes along the path — cure, damage, the linear-comparison composites to come — integrates incrementally. Outputs stress, tangent. Replaces incremental_linear_elasticity and incremental_solid_material; the finite-strain incremental_saint_venant_kirchhoff is left for the Hencky/finite-strain items.

Driver fix found by the report: under stress control the stress passes exactly through zero at a load reversal, and the hypoelastic sum leaves 1e-13 of round-off there; the drivers' purely relative criterion (|P:(σ − σ*)| ≤ tol·max(|σ|, |P:σ*|)) could then never be met. Both drivers now also scale with the step's own stress scale, |C|·|ε − εₙ| (|A|·|F − Fₙ| at finite strain). Covered by UniaxialStressCycleThroughZeroConverges.

Tests (test_incremental_elasticity, 4): equals the total form at every step of a load–unload path with shear and closes the cycle, with one history tensor in STATEV; the uniaxial-stress cycle through zero converges; a repeated trial from the same state does not accumulate; creation from JSON.

Report docs/materials/incremental_linear_elasticity/report.json: the standard cases plus a closed tension–shear–unload path; checks: equals linear_elasticity on every path, E under uniaxial stress, zero stress after the closed path; a step-size study showing the update exact for any step with a constant tangent.

754/754 tests pass (GCC 14, reports on).

JSON input

New material types: incremental_linear_elasticity. A json_model document that loads (the FromJson test of the PR):

{"materials": [
  {"type": "external_strain_source", "name": "strain"},
  {"type": "linear_elasticity", "name": "total", "strain_producer_name": "strain", "K": 100.0, "G": 40.0},
  {"type": "incremental_linear_elasticity", "name": "inc", "strain_source": "strain", "tangent_source": "total"}
]}

tangent_source is any producer of a tangent (here a constant one; a cure- or damage-dependent stiffness is the point of the material); outputs stress (history) and tangent.

[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.

incremental_linear_elasticity — needs changes (small)

• [should] tangent reported as C_{n+1} only; for strain-dependent C the consistent tangent needs (∂C/∂ε):Δε. State "exact for strain-independent C" and add a case where tangent_source actually varies — currently the feature is exercised nowhere (report uses constant total::tangent, test compares against a constant tangent).
• [should] F-driver convergence-scale hunk untested.
• [nit] void update() silently overrides the dead virtual without override (others use compute); unnecessary tensor2{…} temporary; no graph-level revert() test.