Some solvers don't work with ArrayPartition + ExtendedJumpArray

prob = ODEProblem((du, u, p, t) -> du .= 0, ArrayPartition([0.]), (0, 10.))
jump = VariableRateJump((u, p, t) -> 1, int -> (int.u += 1))
jprob = JumpProblem(prob, Direct(), jump)
solve(jprob, KenCarp4())          # can replace this by TRBDF2

yields

BoundsError: attempt to access 1-element ArrayPartition{Float64, Tuple{Vector{Float64}}} at index [2:2]

Stacktrace:
  [1] throw_boundserror(A::ArrayPartition{Float64, Tuple{Vector{Float64}}}, I::Tuple{UnitRange{Int64}})
    @ Base ./abstractarray.jl:691
  [2] checkbounds
    @ ./abstractarray.jl:656 [inlined]
  [3] _setindex!
    @ ./multidimensional.jl:893 [inlined]
  [4] setindex!
    @ ./abstractarray.jl:1315 [inlined]
  [5] _typed_vcat!(a::ArrayPartition{Float64, Tuple{Vector{Float64}}}, V::Tuple{ArrayPartition{Float64, Tuple{Vector{Float64}}}, Vector{Float64}})
    @ Base ./abstractarray.jl:1551
  [6] _typed_vcat
    @ ./abstractarray.jl:1543 [inlined]
  [7] typed_vcat
    @ ./abstractarray.jl:1619 [inlined]
  [8] vcat
    @ ./abstractarray.jl:1535 [inlined]
  [9] zeromatrix(A::ExtendedJumpArray{Float64, 1, ArrayPartition{Float64, Tuple{Vector{Float64}}}, Vector{Float64}})
    @ DiffEqJump ~/.julia/packages/DiffEqJump/x05Qi/src/extended_jump_array.jl:36

The issue seems to be that vec applied to an ArrayPartition returns a tuple, not a vector. Running the above with Vern7 gives

MethodError: no method matching (::DiffEqJump.var"#jump_f#123"{ODEProblem{ArrayPartition{Float64, Tuple{Vector{Float64}}}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, var"#308#309", UniformScaling{Bool}, (...), typeof(SciMLBase.DEFAULT_OBSERVED), Nothing}, Base.Pairs{Symbol, Union{}, Tuple{}, NamedTuple{(), Tuple{}}}, SciMLBase.StandardODEProblem}, JumpSet{Tuple{VariableRateJump{var"#310#312", var"#311#313", Nothing, Float64, Int64}}, Tuple{}, Nothing, Nothing}})(::Vector{Float64}, ::ExtendedJumpArray{Float64, 1, ArrayPartition{Float64, Tuple{Vector{Float64}}}, Vector{Float64}}, ::SciMLBase.NullParameters, ::Float64)
Closest candidates are:
  (::DiffEqJump.var"#jump_f#123")(::ExtendedJumpArray, ::ExtendedJumpArray, ::Any, ::Any) at ~/.julia/packages/DiffEqJump/x05Qi/src/problem.jl:125

Stacktrace:
  [1] (::ODEFunction{true, DiffEqJump.var"#jump_f#123"{ODEProblem{ArrayPartition{Float64, Tuple{Vector{Float64}}}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, var"#308#309", UniformScaling{Bool}, (...), typeof(SciMLBase.DEFAULT_OBSERVED), Nothing}, Base.Pairs{Symbol, Union{}, Tuple{}, NamedTuple{(), Tuple{}}}, SciMLBase.StandardODEProblem}, JumpSet{Tuple{VariableRateJump{var"#310#312", var"#311#313", Nothing, Float64, Int64}}, Tuple{}, Nothing, Nothing}}, UniformScaling{Bool}, (...), typeof(SciMLBase.DEFAULT_OBSERVED), Nothing})(::Vector{Float64}, ::Vararg{Any})
    @ SciMLBase ~/.julia/packages/SciMLBase/pr0Dt/src/scimlfunctions.jl:345
  [2] ode_determine_initdt(u0::ExtendedJumpArray{...}, t::Float64, tdir::Float64, dtmax::Float64, abstol::Float64, reltol::Float64, internalnorm::typeof(DiffEqBase.ODE_DEFAULT_NORM), prob::ODEProblem{...}, ...)
    @ OrdinaryDiffEq ~/.julia/packages/OrdinaryDiffEq/Op0Oq/src/initdt.jl:49

Here the problem is in initdt, providing dt = ... to the solve call bypasses the issue. Sorry for the long error messages; the problem seems to be that ArrayPartition does not follow some standard array interface conventions - any advice for convenient alternatives when solving problems with multiple components?

Reproduce the ArrayPartition example with KenCarp4, TRBDF2, and Vern7. Start by reading src/extended_jump_array.jl, especially zeromatrix, then inspect src/problem.jl and the initdt path named in the stack trace. Done means the reported solver cases run without the BoundsError or MethodError, with regression coverage for the reproduction.