Mark null ODE solutions as dense

We actually know the entire time series for the solution, so we should state that we know that via `dense=true`.

```julia
using OrdinaryDiffEqDefault, ModelingToolkit
using ModelingToolkit: t_nounits as t, D_nounits as D

@variables x(t) y(t)
eqs = [y ~ x, y ~ sin(t)]
@mtkbuild sys = ODESystem(eqs, t)

tspan = (0.0, 10.0)
prob = ODEProblem(sys, [], tspan)
sol = solve(prob)
using Plots
plot(sol, idxs = [x])
savefig("plot.png")
```

Before and after

Author: ChrisRackauckas

src/solve.jl

@@ -748,7 +748,7 @@ function build_null_solution(prob::AbstractDEProblem, args...;
 
     prob, success = hack_null_solution_init(prob)
     retcode = success ? ReturnCode.Success : ReturnCode.InitialFailure
-    build_solution(prob, nothing, ts, timeseries; retcode)
+    build_solution(prob, nothing, ts, timeseries; dense=true, retcode)
 end
 
 function build_null_solution(