fix: biconnected_components() now returns edge and vertex sequences again

[szhorvat]

This is a temporary workaround for #1203 until auto-generation is updated to handle this situation. It restores the 1.2.11 behaviour.

@krlmlr I recommend this for the next bugfix release, but we'll have to find a proper solution later, with auto-generation.

[aviator-app]

Current Aviator status

Aviator will automatically update this comment as the status of the PR changes.
Comment /aviator refresh to force Aviator to re-examine your PR (or learn about other /aviator commands).

This PR was merged using Aviator.

---

See the real-time status of this PR on the Aviator webapp.

Use the Aviator Chrome Extension to see the status of your PR within GitHub.

[krlmlr]

Thanks. We want to move away from using .Call() outside aaa-auto.R . Can we achieve the same by merely wrapping biconnected_components_impl() instead of killing it?

[szhorvat]

I can't, as in I don't know how. The generated version is buggy. This would require a change to stimulus. I could have patched aaa-auto.R directly, but that causes a lot of inconvenience.

[krlmlr]

Can you please add a test? Happy to propose something.

[szhorvat]

I won't have time for a test at the moment, but since this is an obvious bugfix, it would be good to get it out soon ...

Of course eventually we'll need a test. If we had one, this breakage would have been detected.

[szhorvat]

@krlmlr There is already a test for this function. What we need to do is make sure that the components of the result list are named as expected, and are of the correct class (igraph.vs and not numeric). How do you do this? Just check names() and class()?

[krlmlr]

5556282 is what I mean by wrapping. I believe there's always a way, our autogeneration is not that bad.

The other commits are cosmetics and add a test.