Add SIMD RK methods to NonStiffODE benchmarks

This PR adds the SIMD Runge-Kutta methods (MER5v2, MER6v2, RK6v4) from OrdinaryDiffEqSIMDRK
to the existing non-stiff ODE benchmarks for performance comparison.

- Updated FitzhughNagumo_wpd.jmd with SIMD methods
- Updated LotkaVolterra_wpd.jmd with SIMD methods
- Updated RigidBody_wpd.jmd with SIMD methods
- Updated linear_wpd.jmd with SIMD methods
- Added OrdinaryDiffEqSIMDRK to Project.toml dependencies

These methods use SIMD instructions for improved performance on compatible hardware.

Based on SciML/SIMDRungeKutta.jl#16

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>

Author: ChrisRackauckas

benchmarks/NonStiffODE/FitzhughNagumo_wpd.jmd

@@ -8,9 +8,9 @@ author: Chris Rackauckas
 The purpose of this is to see how the errors scale on a standard nonlinear problem.
 
 ```julia
-using OrdinaryDiffEq, ParameterizedFunctions, ODEInterface,
+using OrdinaryDiffEq, OrdinaryDiffEqCore, ParameterizedFunctions, ODEInterface,
       ODEInterfaceDiffEq, LSODA, Sundials, DiffEqDevTools,
-      StaticArrays
+      StaticArrays, OrdinaryDiffEqSIMDRK
 using Plots;
 gr()
 
@@ -46,7 +46,10 @@ setups = [Dict(:alg=>DP5())
           Dict(:alg=>Tsit5())
           Dict(:alg=>Vern6())
           Dict(:alg=>Tsit5(), :prob_choice => 2)
-          Dict(:alg=>Vern6(), :prob_choice => 2)]
+          Dict(:alg=>Vern6(), :prob_choice => 2)
+          Dict(:alg=>MER5v2(), :prob_choice => 2)
+          Dict(:alg=>MER6v2(), :prob_choice => 2)
+          Dict(:alg=>RK6v4(), :prob_choice => 2)]
 wp = WorkPrecisionSet(probs, abstols, reltols, setups; appxsol = test_sol,
     save_everystep = false, numruns = 100, maxiters = 1000)
 plot(wp)

benchmarks/NonStiffODE/LotkaVolterra_wpd.jmd

@@ -8,8 +8,8 @@ author: Chris Rackauckas
 The purpose of this problem is to test the performance on easy problems. Since it's periodic, the error is naturally low, and so most of the difference will come down to startup times and, when measuring the interpolations, the algorithm choices.
 
 ```julia
-using OrdinaryDiffEq, ParameterizedFunctions, ODEInterfaceDiffEq, LSODA,
-      Sundials, DiffEqDevTools, StaticArrays
+using OrdinaryDiffEq, OrdinaryDiffEqCore, ParameterizedFunctions, ODEInterfaceDiffEq, LSODA,
+      Sundials, DiffEqDevTools, StaticArrays, OrdinaryDiffEqSIMDRK
 
 f = @ode_def LotkaVolterra begin
     dx = a*x - b*x*y
@@ -43,7 +43,10 @@ setups = [Dict(:alg=>DP5())
           Dict(:alg=>Tsit5())
           Dict(:alg=>Vern6())
           Dict(:alg=>Tsit5(), :prob_choice => 2)
-          Dict(:alg=>Vern6(), :prob_choice => 2)]
+          Dict(:alg=>Vern6(), :prob_choice => 2)
+          Dict(:alg=>MER5v2(), :prob_choice => 2)
+          Dict(:alg=>MER6v2(), :prob_choice => 2)
+          Dict(:alg=>RK6v4(), :prob_choice => 2)]
 wp = WorkPrecisionSet(probs, abstols, reltols, setups; appxsol = test_sol,
     save_everystep = false, maxiters = 10000, numruns = 100)
 plot(wp)

benchmarks/NonStiffODE/Project.toml

@@ -4,6 +4,8 @@ LSODA = "7f56f5a3-f504-529b-bc02-0b1fe5e64312"
 ODEInterface = "54ca160b-1b9f-5127-a996-1867f4bc2a2c"
 ODEInterfaceDiffEq = "09606e27-ecf5-54fc-bb29-004bd9f985bf"
 OrdinaryDiffEq = "1dea7af3-3e70-54e6-95c3-0bf5283fa5ed"
+OrdinaryDiffEqCore = "bbf590c4-e0c2-4248-924d-a24c5c6facc5"
+OrdinaryDiffEqSIMDRK = "dc97f408-0ac8-40da-bfeb-e1efba43c4c9"
 ParameterizedFunctions = "65888b18-ceab-5e60-b2b9-181511a3b968"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
@@ -17,6 +19,8 @@ LSODA = "0.7"
 ODEInterface = "0.5"
 ODEInterfaceDiffEq = "3.10"
 OrdinaryDiffEq = "6.19"
+OrdinaryDiffEqCore = "1"
+OrdinaryDiffEqSIMDRK = "1"
 ParameterizedFunctions = "5.3"
 Plots = "1.4"
 SciMLBenchmarks = "0.1"

benchmarks/NonStiffODE/RigidBody_wpd.jmd

@@ -4,8 +4,8 @@ author: Chris Rackauckas
 ---
 
 ```julia
-using OrdinaryDiffEq, ParameterizedFunctions, ODEInterfaceDiffEq, LSODA,
-      Sundials, DiffEqDevTools, StaticArrays
+using OrdinaryDiffEq, OrdinaryDiffEqCore, ParameterizedFunctions, ODEInterfaceDiffEq, LSODA,
+      Sundials, DiffEqDevTools, StaticArrays, OrdinaryDiffEqSIMDRK
 
 k(t) = 0.25*sin(t)^2
 
@@ -40,7 +40,10 @@ setups = [Dict(:alg=>DP5())
           Dict(:alg=>Tsit5())
           Dict(:alg=>Vern6())
           Dict(:alg=>Tsit5(), :prob_choice => 2)
-          Dict(:alg=>Vern6(), :prob_choice => 2)]
+          Dict(:alg=>Vern6(), :prob_choice => 2)
+          Dict(:alg=>MER5v2(), :prob_choice => 2)
+          Dict(:alg=>MER6v2(), :prob_choice => 2)
+          Dict(:alg=>RK6v4(), :prob_choice => 2)]
 wp = WorkPrecisionSet(probs, abstols, reltols, setups; appxsol = test_sol,
     save_everystep = true, numruns = 100, maxiters = 10000)
 plot(wp)

benchmarks/NonStiffODE/linear_wpd.jmd

@@ -23,7 +23,7 @@ rapid development has its advantages.
 ## Setup
 
 ```julia
-using OrdinaryDiffEq, Sundials, DiffEqDevTools, Plots, ODEInterfaceDiffEq, LSODA
+using OrdinaryDiffEq, OrdinaryDiffEqCore, Sundials, DiffEqDevTools, Plots, ODEInterfaceDiffEq, LSODA, OrdinaryDiffEqSIMDRK
 using Random
 Random.seed!(123)
 gr()
@@ -105,7 +105,10 @@ algorithms:
 setups = [Dict(:alg=>DP5())
           Dict(:alg=>BS3())
           Dict(:alg=>BS5())
-          Dict(:alg=>Tsit5())]
+          Dict(:alg=>Tsit5())
+          Dict(:alg=>MER5v2())
+          Dict(:alg=>MER6v2())
+          Dict(:alg=>RK6v4())]
 wp = WorkPrecisionSet(prob, abstols, reltols, setups; save_everystep = false, numruns = 100)
 plot(wp)
 ```