Fix documentation fail check

[arnavk23]

Fixes #170

[arnavk23]

@tmigot Whenever you are free, could you start the workflows on this, would like to see if this works or not (also Copilot, maybe could give good suggestions here). Also could you further review the Cannoles pr on this repo of mine.
Thank you!

[Copilot]

Pull request overview

This PR addresses the documentation CI failure reported in #170 by updating the documentation benchmark example to be more robust across environments and by adjusting test environment dependencies.

Changes:

• Update CUTEst problem selection and constrain the docs benchmark to a small, stable subset with error handling.
• Add a fallback benchmark model (ADNLPModels) when CUTEst decoding yields no usable problems during docs build.
• Add DCISolver to test/Project.toml dependencies.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
test/Project.toml	Adds DCISolver to the test environment dependencies.
docs/src/2-benchmark.md	Makes the docs benchmark faster and more resilient (select API update, preferred problem subset, try/catch, and fallback model).

Comments suppressed due to low confidence (2)

docs/src/2-benchmark.md:42

• cutest_problems = Any[] makes the collection type-unstable and can lead to slower execution and poorer dispatch (and may not match downstream method signatures if they expect AbstractNLPModel). Prefer a concretely-typed container (e.g., CUTEstModel[] if you only keep CUTEst models, or AbstractNLPModel[] if you need to mix in the ADNLPModels fallback).

``` @example ex1
using DCISolver, NLPModelsIpopt

**docs/src/2-benchmark.md:103**
* With the new fallback to a toy `ADNLPModel` when `cutest_problems` is empty, the surrounding narrative and later plot titles still describe the benchmark as “on CUTEst”. Consider carrying a label (e.g., `benchmark_name`) that switches when the fallback is used, so the rendered docs don’t become misleading on platforms where CUTEst decoding fails.

pretty_stats(stats[:dcildl])

or it can also be used to make performance profiles.

``` @example ex1
using Plots
gr()

legend = Dict(
  :neval_obj => "number of f evals",
  :neval_cons => "number of c evals",
  :neval_grad => "number of ∇f evals",
  :neval_jac => "number of ∇c evals",
  :neval_jprod => "number of ∇c*v evals",

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[arnavk23]

@tmigot I found the root API change: CUTEst replaced select with select_sif_problems

[tmigot]

the doc still fails