docs: namespace MTK analysis-point helpers and import ImplicitEuler

After #448 unblocked the connections.md operating_point error, the
Documentation build surfaced two further breakages on main:

- `dc_motor_pi.md` uses `Blocks.get_sensitivity`, `Blocks.get_comp_sensitivity`,
  `Blocks.get_looptransfer`. Those helpers moved out of
  `ModelingToolkitStandardLibrary.Blocks` into `ModelingToolkit` proper, so the
  `Blocks.` prefix now hits `UndefVarError`. Use `ModelingToolkit.` instead.
- `input_component.md`'s `custom_component_external_data` example calls
  `solve(prob, ImplicitEuler())`, but OrdinaryDiffEq v7 no longer re-exports
  `ImplicitEuler`. Add `using OrdinaryDiffEqSDIRK: ImplicitEuler`.

Two further docs failures remain (separate root causes, deferred):
- `custom_component.md` `@mtkmodel ChaoticAttractor` can't see the
  `NonlinearResistor` defined in an earlier `@example` block of the same
  name — looks like a hygiene/lookup issue in the SciCompDSL `@mtkmodel`
  expansion when the constructor body runs in a Documenter `@example`
  module.
- `MethodError: fntype_X_Y(::Type{Real})` from SymbolicUtils inside
  the BSplineInterpolation/parametrized_interpolation example —
  upstream SymbolicUtils dispatch gap.

Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>

Author: ChrisRackauckas

docs/src/tutorials/dc_motor_pi.md

@@ -117,12 +117,12 @@ T(s) &= \dfrac{P(s)C(s)}{I + P(s)C(s)}
 using ControlSystemsBase
 # Get sensitivity function
 matrices_S,
-simplified_sys_S = Blocks.get_sensitivity(
+simplified_sys_S = ModelingToolkit.get_sensitivity(
     model, :y, op = Dict(unknowns(sys) .=> 0.0))
 So = ss(matrices_S...) |> minreal # The output-sensitivity function as a StateSpace system
 # Get complementary sensitivity function
 matrices_T,
-simplified_sys_T = Blocks.get_comp_sensitivity(
+simplified_sys_T = ModelingToolkit.get_comp_sensitivity(
     model, :y, op = Dict(unknowns(sys) .=> 0.0))
 To = ss(matrices_T...)# The output complementary sensitivity function as a StateSpace system
 bodeplot([So, To], label = ["S" "T"], plot_title = "Sensitivity functions",
@@ -133,7 +133,7 @@ Similarly, we may compute the loop-transfer function and plot its Nyquist curve
 
 ```@example dc_motor_pi
 matrices_L,
-simplified_sys_L = Blocks.get_looptransfer(
+simplified_sys_L = ModelingToolkit.get_looptransfer(
     model, :y, op = Dict(unknowns(sys) .=> 0.0))
 L = -ss(matrices_L...) # The loop-transfer function as a StateSpace system. The negative sign is to negate the built-in negative feedback
 Ms, ωMs = hinfnorm(So) # Compute the peak of the sensitivity function to draw a circle in the Nyquist plot

docs/src/tutorials/input_component.md

@@ -187,6 +187,7 @@ using ModelingToolkit
 using ModelingToolkit: t_nounits as t, D_nounits as D
 using ModelingToolkitStandardLibrary.Blocks
 using OrdinaryDiffEq
+using OrdinaryDiffEqSDIRK: ImplicitEuler
 
 const rdata = Ref{Vector{Float64}}()