[Doc] review of examples published on the website

[j-l-s]

OptimalControlProblems Website Examples Review

Report on macOS-ARM (M1)

Tested from a fresh environment (rm -fr $HOME/.julia).
Julia version: 1.12.5 (Julia 1.12.5+0.aarch64.apple.darwin14).
Ran all examples from https://control-toolbox.github.io/OptimalControlProblems.jl/stable/

Global Remarks

Erroneous Comment (All examples except beam of course)

First line comment is incorrect:

using OptimalControlProblems  # to access the Beam model

Console Output Formatting

Third cell output (e.g., metadata(:beam)[:parameters]) lacks website formatting:

julia> metadata(:beam)[:parameters]
(t0 = 0, tf = 1, x₁_l = 0, x₁_u = 0.1, x₁_t0 = 0, x₂_t0 = 1, x₁_tf = 0, x₂_tf = -1)

Website shows prettier named tuple

Plot Label Differences (Final Plots)

Local	Website
No "JuMP" in labels (red curves plotted OK)	Single global label (top-left)
Per-subplot labels	One global label

Plotting loop issue? Cell 3 and final cell code is/looks similar....

Package Installation

Adding OptimalControlProblems

(@v1.12) pkg> add OptimalControlProblems
# ...
⌅ [54762871] + CTBase v0.16.2
⌅ [790bbbee] + CTDirect v0.17.4
⌅ [34c4fa32] + CTModels v0.6.9
⌅ [be4d8f0f] + Quadmath v0.5.13

# ⌃/⌅: New versions available; ⌅ blocked by compatibility constraints.

Why aren't CT* packages up-to-date? then installing OptimalControlProblems on a fresh/empty environment ?

macOS-Specific Warning

(@v1.12) pkg> add NLPModelsIpopt
# ...
┌ Warning: MKL is not available/installed.
└ @ MKL ~/.julia/packages/MKL/JVoAz/src/MKL.jl:39

MKL unavailable on ARM/macOS.[1]

Minor Numerical Differences

Small discrepancies versus website results (optimal values). Likely due to ARM vs Intel floating-point behavior, I guess