[ChrisRackauckas-Claude]

What changed and why

Bump EasyModelAnalysis from 1.3.0 to the next consecutive minor release, 1.4.0. This publishes the documented public-surface restoration merged in #327; the minor bump matches the policy for new public API in a package at major version 1.

Ignore this PR until it has been reviewed by @ChrisRackauckas.

Verification

Run locally on the exact release commit with Julia 1.12.6:

GROUP=QA julia --project -e 'using Pkg; Pkg.test()'
Test Summary: | Pass  Total      Time
QA/qa.jl      |   21     21  14m41.0s
Testing EasyModelAnalysis tests passed

julia --project -e 'using Pkg; Pkg.test()'
Core/basics.jl              9/9    18m13.2s
Core/ensemble.jl            1/1    31m04.1s
Core/examples.jl            4/4         7.8s
Core/precompile_workload.jl 1/1         0.2s
Core/sensitivity.jl         4/4    72m10.6s
Core/threshold.jl          12/12    7m17.8s
Datafit/datafit.jl         10/10   12m09.3s
Testing EasyModelAnalysis tests passed

julia --project=@runic -e 'using Runic; exit(Runic.main(["--check", "."]))'
exit 0

git diff --check
exit 0

git diff --unified=0 | typos --format brief -
exit 0

The docs build and downstream/allowed-to-fail jobs were not run locally because this PR changes only the package version.

CI note: Documentation fails during Pkg.develop/resolution with the Catlab/DataStructures/SymbolicUtils incompatibility recorded in the handoff as pre-existing: https://github.com/SciML/EasyModelAnalysis.jl/actions/runs/32902642041/job/97979687942. This version-only PR remains draft while that clean-main classification is independently re-audited.

AI agent continuation: OpenAI Codex
Codex-Session-ID: 01a03a11-47c2-77e0-858b-167004f0b79c

🤖 Generated with Claude Code

https://claude.ai/code/session_01Rmh6B9eoW3N8oCbuuVPVxJ

[ChrisRackauckas-Claude]

Clean-main investigation confirms this release PR is blocked by a real pre-existing docs migration, not by the version bump.

• Current main at 06718cc5fa846b46d0575875e124c192e2e5b448 reproduces the resolver failure on Julia 1.12.7.
• The first bad EasyModelAnalysis commit is a363934cbcc4ade3e7e851cb80201023900ff78b: a363934
• Its predecessor resolves successfully; the first-bad commit removed ModelingToolkit 9 compatibility while current upstream constraints still deadlock around AlgebraicPetri and GeneralizedGenerated.
• An experimental four-upstream compat overlay resolved and precompiled with ModelingToolkit 11.40.0, Catalyst 16.3.0, OrdinaryDiffEq 7.7.0, DataStructures 0.19.6, AlgebraicPetri 0.10.0, and GeneralizedGenerated 0.3.3. This proves one exact resolution tuple, not minimal compatible floors.
• Full docs under that overlay still produced 29 executed-example failures across five pages, including ReactionSystem conversion, uncertainty indexing, parameter-only events, and solving a vector of ODEProblems. The process remained CPU-active until the four-hour limit and exited 124.

A compat-only edit is therefore not an honest fix. The remediation needs coordinated GeneralizedGenerated/AlgebraicPetri support plus a focused EasyModelAnalysis migration. No tests or docs were disabled, and no external PR was opened without permission.

Failing job: https://github.com/SciML/EasyModelAnalysis.jl/actions/runs/32902642041/job/97979687942
First-bad PR and matching docs failure: #320 and https://github.com/SciML/EasyModelAnalysis.jl/actions/runs/31843613416/job/94905485696

AI-Agent: OpenAI Codex
Codex-Session-ID: 01a03a11-47c2-77e0-858b-167004f0b79c