Add stress_from_state material-specific implementations for MaterialModelsBase.jl#12

[knutambot]

Summary

• Introduces calculate_current_stress(m, ϵ, state), prototyping MaterialModelsBase.jl#12: given a strain and an already-converged/current material state, it returns the corresponding stress without invoking any local Newton iteration that would advance history variables. Useful for postprocessing, e.g. recomputing stress at a strain that differs slightly from the one used to produce state.
• Implemented for LinearElastic (via a generic NoMaterialState fallback), Plastic, GeneralizedMaxwell, and the RotatedMaterial wrapper.
• Supports lower-dimensional stress from a lower-dimensional strain via MaterialModelsBase.ReducedStressState, for LinearElastic and Plastic. This replaces the ad hoc, per-material calculate_stress dispatch hack used for plane-stress postprocessing in FerriteAssembly.jl#94's mixed_materials tutorial (that tutorial is not changed here — a follow-up PR should call this function instead).

Review process

Went through two rounds of independent Codex review (same thread):

• Plan review caught two real bugs before implementation: (1) the reduced-dimensional Plastic formula originally reduced the plastic strain before subtracting it, which discards its out-of-plane component and gives the wrong answer for PlaneStrain (fixed by expanding the total strain to 3d first, then subtracting the full 3d plastic strain); (2) the RotatedMaterial/ReducedStressState wrapper methods were ambiguous with the NoMaterialState fallback when wrapping a stateless material (fixed with explicit disambiguating overloads).
• Diff review against the finished implementation: no further findings.

Test plan

• Pkg.test() passes: all pre-existing tests plus 16 new tests in test/test_current_stress.jl, including "frozen-state" checks that evaluating at a different strain gives a purely elastic increment rather than triggering a fresh plastic/viscous correction.
• docs/make.jl builds cleanly (only pre-existing, unrelated warnings).
• This package has no literate tutorials/howtos, so that step doesn't apply here.

Remaining risks / deliberately out of scope

• Reduced-dimensional support for GeneralizedMaxwell and for RotatedMaterial wrapping a stateful material is not implemented (calling calculate_current_stress on those combinations throws a clean MethodError rather than silently giving a wrong answer).
• FiniteStrainPlastic, CrystalPlasticity, and the hyperelastic models are not covered, since the motivating use case (FerriteAssembly.jl#94's plasticity postprocessing) is small-strain.

🤖 Generated with Claude Code

https://claude.ai/code/session_0181fqMcFarfUzBir46jb2rp

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.18%. Comparing base (7c09cff) to head (501a024).

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #13      +/-   ##
==========================================
+ Coverage   96.10%   96.18%   +0.07%     
==========================================
  Files          19       20       +1     
  Lines         745      760      +15     
==========================================
+ Hits          716      731      +15     
  Misses         29       29              

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[KnutAM]

Suggested change

	function calculate_current_stress(stress_state::MMB.AbstractStressState, m::FiniteStrainPlastic, F, state::FiniteStrainPlasticState)
	frozen = FrozenStressMaterial(F_ -> calculate_PKstress(m, state, F_))
	σ, _, _, _ = MMB.material_response(stress_state, frozen, F, MMB.NoMaterialState{eltype(F)}())
	return σ
	end
	function calculate_current_stress(stress_state::MMB.AbstractStressState, m::AbstractMaterial, strain, state::AbstractMaterialState)
	frozen = FrozenStressMaterial(e -> calculate_current_stress(m, state, e))
	σ, _, _, _ = MMB.material_response(stress_state, frozen, strain, MMB.NoMaterialState{eltype(F)}())
	return σ
	end

[KnutAM]

Suggested change

	function MMB.material_response(fm::FrozenStressMaterial, F::Tensor{2,3}, old::MMB.AbstractMaterialState, args...)
	function MMB.material_response(fm::FrozenStressMaterial, F::SecondOrderTensor{3}, old::MMB.AbstractMaterialState, args::Vararg{N}) where {N}

[KnutAM]

Can be removed, since we want to special case to avoid calculating the gradient in this case.

[knutambot]

Addressed in 06f737d: added a generic calculate_current_stress(stress_state, m::AbstractMaterial, strain, state::AbstractMaterialState) fallback that wraps a model's own full-dimensional calculate_current_stress in FrozenStressMaterial and rides MaterialModelsBase's existing stress-state Newton iteration, autodiff-ing through it for the tangent — so a ReducedStressState now "just works" for any material once its full-dimensional method exists (the FiniteStrainPlastic-specific reduced method is gone, replaced by this).

Also removed the full-3d generic NoMaterialState fallback and replaced it with dedicated gradient-free methods for LinearElastic and AbstractHyperElastic — material_response was computing (and calculate_current_stress discarding) an unnecessary tangent for the hyperelastic models via compute_stress_and_tangent's second differentiation.

Bonus from the generalization: GeneralizedMaxwell and RotatedMaterial wrapping a stateful material now get ReducedStressState support for free too (previously unsupported) — added tests for both.

🤖 Generated with Claude Code

[knutambot]

Migrated to build on top of the upstreamed interface: MaterialModelsBase.jl#21 now provides stress_from_state (the renamed, generalized version of what was prototyped here), including the generic machinery (FrozenStressMaterial, the generic reduced-dimensional fallback, ReducedStressState delegation).

This PR now keeps only the material-specific implementations, extending MaterialModelsBase.stress_from_state:

• LinearElastic/AbstractHyperElastic: gradient-free NoMaterialState overrides.
• Plastic: full + reduced-dim analytic fast path.
• GeneralizedMaxwell, FiniteStrainPlastic: full-dimensional only — reduced-dim now comes for free from upstream's generic fallback.
• RotatedMaterial: full-dimensional wrapper.

⚠️ Temporary, must revert before merging MMB#21: Project.toml and docs/Project.toml currently pin MaterialModelsBase via [sources] to knutambot/MaterialModelsBase.jl#cb/calculate_current_stress (unmerged), so this branch can build/test/doc against the new interface ahead of that PR merging, per your request. Both entries are commented # TEMPORARY. Once MMB#21 merges and releases, these need reverting to the registry-tracked package and the [compat] bound tightened to exclude older releases lacking stress_from_state.

Pkg.test() and docs/make.jl both pass against that branch.

🤖 Generated with Claude Code

[KnutAM]

Suggested change

	@test state1.Fp != state0.Fp # sanity: this test only matters if plastic loading occurred
	@assert state1.Fp !≈ state0.Fp # sanity: this test only matters if plastic loading occurred

[KnutAM]

Suggested change

	@test state2_true.Fp != state1.Fp
	@assert state2_true.Fp !≈ state1.Fp