Add downgrade CI for all sublibraries

[ChrisRackauckas]

Summary

• Adds comprehensive downgrade CI testing for all sublibraries
• Tests the main package (.) and all 6 sublibraries in
• Each sublibrary is tested individually with downgraded dependencies
• Uses julia-downgrade-compat@v2 with ALLOW_RERESOLVE: false

Sublibraries Covered

Main package:

6 Sublibraries:

Benefits

• Ensures compatibility across the entire package ecosystem
• Catches dependency issues early for each sublibrary
• Provides comprehensive downgrade testing coverage
• Maintains lean testing with focused sublibrary-specific tests

Test plan

• Verify workflow triggers correctly on PRs and pushes
• Confirm each sublibrary builds and tests successfully
• Check that downgrade testing works for all projects
• Validate matrix strategy runs all combinations

🤖 Generated with Claude Code

[github-actions]

Benchmark Results

Click to check benchmark results

	master	36d9c95...	master / 36d9c95...
Simple Pendulum/IIP/BoundaryValueDiffEqMIRK.MIRK2()	0.6 ± 0.02 s	0.592 ± 0.017 s	1.01 ± 0.044
Simple Pendulum/IIP/BoundaryValueDiffEqMIRK.MIRK3()	13.7 ± 0.57 ms	13.7 ± 0.52 ms	1 ± 0.056
Simple Pendulum/IIP/BoundaryValueDiffEqMIRK.MIRK4()	3.16 ± 0.17 ms	3.13 ± 0.22 ms	1.01 ± 0.089
Simple Pendulum/IIP/BoundaryValueDiffEqMIRK.MIRK5()	9.35 ± 1.2 ms	9.22 ± 1.3 ms	1.01 ± 0.2
Simple Pendulum/IIP/BoundaryValueDiffEqMIRK.MIRK6()	1.63 ± 0.3 ms	1.62 ± 0.28 ms	1.01 ± 0.26
Simple Pendulum/IIP/MultipleShooting(10, Tsit5; grid_coarsening = false)	1.87 ± 0.64 ms	1.83 ± 0.64 ms	1.02 ± 0.5
Simple Pendulum/IIP/MultipleShooting(10, Tsit5; grid_coarsening = true)	3.14 ± 0.98 ms	3.09 ± 0.98 ms	1.01 ± 0.45
Simple Pendulum/IIP/MultipleShooting(100, Tsit5; grid_coarsening = false)	0.0668 ± 0.007 s	0.0672 ± 0.0067 s	0.993 ± 0.14
Simple Pendulum/IIP/MultipleShooting(100, Tsit5; grid_coarsening = true)	0.0816 ± 0.012 s	0.0789 ± 0.014 s	1.03 ± 0.24
Simple Pendulum/IIP/Shooting(Tsit5())	0.239 ± 0.075 ms	0.236 ± 0.076 ms	1.01 ± 0.46
Simple Pendulum/OOP/BoundaryValueDiffEqMIRK.MIRK2()	0.761 ± 0.012 s	0.746 ± 0.011 s	1.02 ± 0.022
Simple Pendulum/OOP/BoundaryValueDiffEqMIRK.MIRK3()	16.7 ± 6.1 ms	16.5 ± 4.8 ms	1.02 ± 0.48
Simple Pendulum/OOP/BoundaryValueDiffEqMIRK.MIRK4()	3.64 ± 0.18 ms	3.58 ± 0.15 ms	1.02 ± 0.065
Simple Pendulum/OOP/BoundaryValueDiffEqMIRK.MIRK5()	10.9 ± 1.2 ms	10.7 ± 0.66 ms	1.03 ± 0.13
Simple Pendulum/OOP/BoundaryValueDiffEqMIRK.MIRK6()	1.88 ± 0.14 ms	1.87 ± 0.13 ms	1 ± 0.1
Simple Pendulum/OOP/MultipleShooting(10, Tsit5; grid_coarsening = false)	3.62 ± 2.9 ms	3.48 ± 2.9 ms	1.04 ± 1.2
Simple Pendulum/OOP/MultipleShooting(10, Tsit5; grid_coarsening = true)	6.1 ± 4.9 ms	5.91 ± 4.9 ms	1.03 ± 1.2
Simple Pendulum/OOP/MultipleShooting(100, Tsit5; grid_coarsening = false)	0.121 ± 0.0073 s	0.116 ± 0.0034 s	1.04 ± 0.07
Simple Pendulum/OOP/MultipleShooting(100, Tsit5; grid_coarsening = true)	0.147 ± 0.0061 s	0.139 ± 0.0062 s	1.05 ± 0.064
Simple Pendulum/OOP/Shooting(Tsit5())	0.635 ± 0.045 ms	0.631 ± 0.038 ms	1.01 ± 0.094
time_to_load	5.42 ± 0.049 s	5.33 ± 0.013 s	1.02 ± 0.0096

### Benchmark Plots A plot of the benchmark results has been uploaded as an artifact to the workflow run for this PR. Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).