[Krastanov-agent]

Summary

• replace the BlossomV runtime install in tests with a deterministic brute-force MWPM oracle using StableRNGs.jl
• keep JET out of the default test environment, run the JET job on Julia 1, pin the JET test install to a Julia-1.12-compatible release, and scope reports to the public package module
• restore the docs job with a minimal Documenter site
• fix current workflow drift: macOS x64 setup-julia v3, missing typos config, changelog skips for dependency PR labels, and benchmark setup including an explicit ASV script path

Verification

• julia --project=. -e 'import Pkg; Pkg.test()'
• JET_TEST=true julia --project=. -e 'import Pkg; Pkg.test()'
• julia --project=docs -e 'import Pkg; Pkg.develop(Pkg.PackageSpec(path=pwd())); Pkg.instantiate(); include("docs/make.jl")'
• typos . --config .typos.toml
• temporary benchmark env loads benchmark/benchmarks.jl and defines SUITE
• git diff --check

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.95%. Comparing base (42c6936) to head (3068aab).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #27      +/-   ##
==========================================
+ Coverage   80.00%   80.95%   +0.95%     
==========================================
  Files           1        1              
  Lines          20       21       +1     
==========================================
+ Hits           16       17       +1     
  Misses          4        4              

☔ 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.

[Krastanov-agent]

Note on the remaining benchmark failure: this workflow runs on pull_request_target, so this PR is evaluated with the benchmark workflow/script state from the current main branch. main does not yet contain benchmark/benchmarks.jl, so AirspeedVelocity fails before it can use the benchmark suite added in this PR. After this PR is merged, future benchmark runs will have the benchmark script available.

[Krastanov-agent]

Benchmark is still failing for the same pull_request_target reason: the check is using the workflow file from base main, not this PR branch. The latest log shows the ASV action still assembles empty script and extra-pkgs inputs, then falls back to /home/runner/.julia/packages/LEMONGraphs/.../benchmark/benchmarks.jl, which does not exist on the current main/released tree.

This PR now adds benchmark/benchmarks.jl and updates .github/workflows/benchmark.yml to pass script: benchmark/benchmarks.jl and extra-pkgs: Graphs. Those workflow changes will take effect for benchmark runs after the PR is merged into main; they cannot affect this PR's own pull_request_target run.