Iris from clique cover allows the use of iris variants

[AlexandreAmice]

This change is 

[AlexandreAmice]

+@cohnt for feature review.

cc @wernerpe if you want to chime in

Reviewable status: LGTM missing from assignee cohnt, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits), missing label for release notes (waiting on @AlexandreAmice)

[AlexandreAmice]

@cohnt If you could suggest fixes to the docs. I don't really have time to comb through and explain the new usage. Also you would know better than me the pitfalls.

Also, there is a footgun in this sample PR where the subspace machinery almost certainly doesn't work since we don't sample from the subspace.

Reviewable status: LGTM missing from assignee cohnt, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits), missing label for release notes (waiting on @AlexandreAmice)

[SeanCurtis-TRI]

planning/iris/iris_from_clique_cover.cc line 246 at r1 (raw file):

    checker.UpdatePositions(clique_ellipse.center(), builder_id);
    log()->debug("Iris builder thread {} is constructing a set.", builder_id);
    std::visit(

BTW You'll want to look at common/overloaded.h and generally such for overloaded to see how we handle variants in drake.

[cohnt]

Reviewed 6 of 6 files at r1, all commit messages.
Reviewable status: 3 unresolved discussions, LGTM missing from assignee cohnt, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits), missing label for release notes (waiting on @AlexandreAmice)

---

planning/iris/iris_from_clique_cover.h line 40 at r1 (raw file):

   * allow more than 1 thread, then the debug visualizations of internal Iris
   * calls will be disabled. This is due to a limitation of drawing to meshcat
   * from outside the main thread.

I would add something like

@throws if the user specifies a parameterization in IrisZoOptions or IrisNp2Options.

Then add a check in IrisFromCliqueCover like the following:

if (!(options.parameterization.get_parameterization()(VectorXd::Zero(ndim)) == VectorXd::Zero(ndim))) {
  throw std::runtime_error("IrisFromCliqueCover does not yet support growing regions on a parameterized subspace.");
}

A test would be something like this test from IrisNp2:

VectorX<Variable> varable_vector(1);
VectorX<Expression> expression_vector(1);
expression_vector[0] = varable_vector[0] + 1;
options.parameterization =
    IrisParameterizationFunction(expression_vector, varable_vector);
DRAKE_EXPECT_THROWS_MESSAGE(
    IrisNp2(*sgcc_ptr, starting_ellipsoid_, domain_, options),
    ".*parameterized subspace.*");
options.parameterization = IrisParameterizationFunction();

And also a

// TODO(cohnt): Support user-specified parameterizations contained in IrisZoOptions and IrisNp2Options.

---

planning/iris/iris_from_clique_cover.cc line 441 at r1 (raw file):

    // Show the samples used in build cliques. Debugging visualization.
    auto meshcat_ptr = std::visit(

btw This snippet appears twice, and could conceivably be needed again elsewhere? Maybe an internal linkage function like GetMeshcatPtrFromOptions (or similarly-named) would be nice.