Fix possible StackOverflowErrors for remake_buffer & InitializationProblems

[SebastianM-C]

For InitializationProblems state_values(prob) can return nothing,
so calls like

remake_buffer(prob, state_values(prob), keys(u0), values(u0))

would lead to StackOverflowErrors due to the fallback for the deprecated Dict method.

function remake_buffer(sys, oldbuffer, idxs, vals)
    remake_buffer(sys, oldbuffer, Dict(idxs .=> vals))
end

This can crash julia if depwarns are enabled. The output would look like

julia> remake(prob1, u0=prob1.u0)
┌ Warning: `remake_buffer(sys, oldbuffer, vals::Dict)` is deprecated, use `remake_buffer(sys, oldbuffer, keys(vals), values(vals))` instead.
│   caller = remake_buffer(sys::NonlinearSystem, oldbuffer::Nothing, idxs::Vector{Int64}, vals::Vector{Any}) at remake.jl:59
└ @ SymbolicIndexingInterface ~/.julia/packages/SymbolicIndexingInterface/GKI0D/src/remake.jl:59
┌ Warning: `remake_buffer(sys, oldbuffer, vals::Dict)` is deprecated, use `remake_buffer(sys, oldbuffer, keys(vals), values(vals))` instead.
│   caller = remake_buffer(sys::NonlinearSystem, oldbuffer::Nothing, idxs::Base.KeySet{Any, Dict{Any, Any}}, vals::Base.ValueIterator{Dict{Any, Any}}) at remake.jl:59
└ @ SymbolicIndexingInterface ~/.julia/packages/SymbolicIndexingInterface/GKI0D/src/remake.jl:59

[2462463] signal 6 (-6): Aborted
in expression starting at none:0
Aborted (core dumped)

Checklist

• Appropriate tests were added
• Any code changes were done in a way that does not break public API
• All documentation related to code changes were updated
• The new code follows the
  contributor guidelines, in particular the SciML Style Guide and
  COLPRAC.
• Any new documentation only uses public API

Additional context

I got the above by running

using ModelingToolkit, OrdinaryDiffEqTsit5
using SciMLBase

function experiment1_sys()
    D = ModelingToolkit.D_nounits
    t = ModelingToolkit.t_nounits
    @parameters p1 = 0.5 [tunable = true] (p23[1:2] = [1, 3.0]) [tunable = true] p4 = 3 * p1 [tunable = false] y0 = 1.2 [tunable = true]
    @variables x(t) = 2p1 y(t) = y0 z(t) = x + y

    eqs = [D(x) ~ p1 * x - p23[1] * x * y
        D(y) ~ -p23[2] * y + p4 * x * y
        z ~ x + y]

    structural_simplify(ODESystem(eqs, t, tspan=(0, 3.0), name=:sys))
end

sys1 = experiment1_sys()
prob1 = ODEProblem(sys1, [sys1.p23 => [2, 4.0]])

remake(prob1, u0=prob1.u0)

on julia 1.11.0 (I tested on Linux so far) with --depwarn=yes --check-bounds=yes

[AayushSabharwal]

The version doesn't need to be bumped, it was done twice on master