added note for the triangle inequality case in TSP

[Schefflera-Arboricola]

Closes #6748

Added the following note in the docstring of traveling_salesman_problem() function:

(Note: If the graph contains any triangle inequalities then it might return a minimal 
approximate cycle with repeating nodes(other than the starting node).)

should we also put a warning message here? because TSP makes sense only for valid graphs, without triangle inequalities. Please let me know what you think.

Thank you :)

[rossbar]

I took the liberty of refactoring the suggestion so that it satisfies both the docstring standard and the formatting rules.

[dschult]

We should also include the case where the graph is not a complete graph (where the triangle inequality definitely doesn't apply -- non-existing edges are length inf so definitely violate the inequality). Perhaps add a phrase like: e.g. when a graph is not a complete graph.

And perhaps this should appear at the end of the paragraph at line 222 instead of in a separate Notes section.

[Schefflera-Arboricola]

@dschult Thank you for the review. I've made the changes. Please let me know if this looks good.

[dschult]

This looks good! Thanks!

[dschult]

To get this to pass the documentation build tests, we need to include a change to the testing system that has laready been merged into upstream main branch. If you want an exercise in git, you can try to "rebase" this PR onto the current main branch. That means you get git to apply the commits from this PR to the current main branch and make this branch point to the new "rebased" code.

Sometimes rebasing is as easy as: git pull upstream main --rebase.
But if the upstream main changes lines that you have also changed, you'll have to update the commits in this PR to work with applying them to the new main branch. That is called "resolving the conflicts". Sometimes that is quite time consuming. I think this one will be "easy".

But I'm confident that we can just merge this and the documentation will work. This PR only changed some words in the docs -- not it's structure or format. :)

[rossbar]

Excellent, thanks @Schefflera-Arboricola for the updates and @dschult for the review & suggestions!