`subset_tunables` seems to modify the system in-place

[SebastianM-C]

Describe the bug 🐞

The original system appears modified after using subset_tunables. As I understand this operation should be out of place. The modification is that tunable_parameters returns the reduced set of parameters instead of the original one.

Expected behavior

The output of tunable_parameters should not change after using subset_tunables.

Minimal Reproducible Example 👇

using ModelingToolkit
using ModelingToolkit: D_nounits as D, t_nounits as t

@parameters α = 1.5 β = 1.0 γ = 3.0 δ = 1.0
@variables x(t) y(t)

eqs = [D(x) ~ α * x - β * x * y,
    D(y) ~ -γ * y + δ * x * y]

model = System(eqs, t, name=:sys)

sys = mtkcompile(model)

tunable_parameters(sys) # 8-element Vector{Any}

subset_tunables(sys, [α])

tunable_parameters(sys) # 5-element Vector{Any}

Error & Stacktrace ⚠️

the first call to tunable_parameters shows

8-element Vector{Any}:
 α
 β
 δ
 γ
 Initial(x(t))
 Initial(yˍt(t))
 Initial(xˍt(t))
 Initial(y(t))

the second

5-element Vector{Any}:
 α
 Initial(x(t))
 Initial(yˍt(t))
 Initial(xˍt(t))
 Initial(y(t))

Environment (please complete the following information):

• Output of using Pkg; Pkg.status()

ModelingToolkit v10.22.0

• Output of using Pkg; Pkg.status(; mode = PKGMODE_MANIFEST)

• Output of versioninfo()

Additional context

Add any other context about the problem here.