work towards MTKV10

[oameye]

This PR needs SciML/ModelingToolkit.jl#3781 and https://github.com/qojulia/SecondQuantizedAlgebra.jl

SciML/ModelingToolkit.jl#3789

[oameye]

Only one test failure left:

order = 2 #order of the cumulant expansion
@cnumbers N Δ g κ Γ R ν M

hc = FockSpace(:cavity)
ha = NLevelSpace(:atom, 2)

h = hc ⊗ ha

k = Index(h, :k, N, ha)
l = Index(h, :l, N, ha)
m = Index(h, :m, N, ha)
n = Index(h, :n, N, ha)

@qnumbers a::Destroy(h)

σ(i, j, k) = IndexedOperator(Transition(h, :σ, i, j), k)
σn(i, j, k) = NumberedOperator(Transition(h, :σ, i, j), k)

# Define the Hamiltonian
H = -Δ*a'a + g*(Σ(a'*σ(1, 2, k), k) + Σ(a*σ(2, 1, k), k))

J = [a, σ(1, 2, l), σ(2, 1, l), σ(2, 2, l)]
rates = [κ, Γ, R, ν]
ops = [a'*a, σ(2, 2, m)]
eqs = meanfield(ops, H, J; rates = rates, order = order)

# System parameters
N_ = 2e5
Γ_ = 1.0 #Γ=1mHz
Δ_ = 2500Γ_ #Δ=2.5Hz
g_ = 1000Γ_ #g=1Hz
κ_ = 5e6*Γ_ #κ=5kHz
R_ = 1000Γ_ #R=1Hz
ν_ = 1000Γ_ #ν=1Hz

ps = [N, Δ, g, κ, Γ, R, ν]
p0 = [N_, Δ_, g_, κ_, Γ_, R_, ν_]

op1 = σ(1, 2, m)
op2 = σ(2, 1, n)

# Test system with no filter func to evaluate corr function

eqs_c_2 = complete(eqs);
eqs_ev = evaluate(eqs_c_2; limits = (N=>3))

u0_ev = zeros(ComplexF64, length(eqs_ev))

@named sys_ev = ODESystem(eqs_ev)
prob_ev = ODEProblem(sys_ev, u0_ev, (0.0, 0.1/50Γ_), ps .=> p0);
sol_ev = solve(prob_ev, Tsit5(), maxiters = 1e7)

corr3 = CorrelationFunction(op1, op2, eqs_c_2)
corr3_ev = evaluate(corr3, 1, 2; limits = (N=>3));

@named csys3 = ODESystem(corr3_ev)

p0_c3 = correlation_p0(corr3_ev, sol_ev.u[end], ps .=> p0)
u0_c3 = correlation_u0(corr3_ev, sol_ev.u[end])
dict = merge(Dict(u0_c3), Dict(p0_c3))

# "var\"⟨σ122⟩\"(τ)" ∈ string.(unknowns(csys3))
# "var\"⟨σ122⟩\"(τ)" ∈ string.(first.(u0_c3))

prob_c3 = ODEProblem(csys3, dict, (0.0, 0.05))
sol_c3 = solve(prob_c3, Tsit5(); saveat = 0.001, maxiters = 1e8);

[codecov]

Codecov Report

❌ Patch coverage is 93.75000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 78.78%. Comparing base (b99c9ae) to head (8d2db63).
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
src/measurement_backaction_indices.jl	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #243      +/-   ##
==========================================
- Coverage   79.38%   78.78%   -0.61%     
==========================================
  Files          17       17              
  Lines        3235     3214      -21     
==========================================
- Hits         2568     2532      -36     
- Misses        667      682      +15     

☔ View full report in Codecov by Sentry.
📢 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.

[oameye]

Only two examples in the docs remain to be fixed: unique_squeezing.md and ramsey_spectroscopy.md.

@ChristophHotter Can you have a look at them?

[oameye]

Migration to new ModelingToolkit.jl API:

• Replaces all instances of ODESystem with System in documentation, markdown examples, and Jupyter notebooks, reflecting the updated ModelingToolkit.jl API. Updates related explanatory text accordingly. [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

• Updates all example code to use the new approach for passing initial conditions and parameters to ODEProblem, using dictionaries and merging as required by the latest ModelingToolkit.jl conventions. [1] [2] [3] [4] [5] [6] [7]

Documentation and example improvements:

• Updates explanatory text in markdown and notebooks to refer to System instead of ODESystem, and clarifies the new parameter/initial condition handling. [1] [2] [3] [4] [5] [6]

Dependency and CI configuration updates:

• Updates the SymbolicUtils dependency in docs/Project.toml to a newer version for compatibility.
• Comments out the 'pre' Julia version in the test matrix and disables some example documentation pages that may not be compatible with the latest changes. [1] [2]
• Updates the Julia version in the documentation deployment workflow from 'lts' to '1'.

[Copilot]

Pull Request Overview

This PR updates QuantumCumulants.jl to work with ModelingToolkit v10 (MTKV10), implementing breaking changes in the API. The main change is replacing ODESystem with System and updating the corresponding function signatures and parameter handling throughout the codebase.

• Replace ODESystem constructor with System across all files
• Update ODEProblem creation to use dict-based parameter/initial condition merging instead of separate parameter lists
• Refactor internal functions to support the new ModelingToolkit API

Reviewed Changes

Copilot reviewed 56 out of 57 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
Multiple test files	Update from ODESystem to System and change ODEProblem parameter syntax
src/diffeq.jl	Major refactoring of system creation logic and parameter extraction
src/correlation.jl	Update correlation function system creation
src/measurement_backaction_indices.jl	Replace ODESystem with System and move utility function
src/utils.jl	Add utility function for averaging operations
docs/ files	Update documentation examples to use new syntax

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[ChristophHotter]

Looks good! Thanks for your work!
I will fix adapt the jupyter notebooks and look at the issue with SQA v0.4.2 later.

[ChristophHotter]

Fix to 0.4.1 for now.
I will look at v0.4.2 later.

[ChristophHotter]

The dict for the ODEProblem needs to be added also for the jupyter-notebooks. I will do it in a separate PR later.

[ChristophHotter]

Why does the function need to be global?

[oameye]

I doesn't, think. I remnant of when I was trying to debug

[oameye]

Can we do SQA 4.2 in a seperate PR?

[oameye]

We should implement a Literate.jl workflow such that the jupyter notebooks should not be updated manually. I propose we merge and do this in a seperate PR.