test cases for IncrementalFixedLagSmoother

[truher]

two cases here:

• SEGV while marginalizing a variable with only a prior.
• IndexError during marginalization that seems distantly caused by the behavior of update()?

[ProfFan]

Thanks! The IncrementalFixedLagSmoother has been known to have problems. There are several workarounds in the issues section, but none of us have the bandwidth for the bugfix right now.

[truher]

no worries, i appreciate your quick response. i can avoid these particular issues, though i'm concerned that there may be more issues lurking in the future. is there a better choice than IncrementalFixedLagSmoother ? The "batch" version doesn't seem to have implemented the marginalCovariance method, which I kinda want.

[dellaert]

I appreciate the tests and might take a look. I have been seeing some other segfaults with 3.12.
What version of python are you using?

[truher]

thank you! i know you're very busy.

re: python version, i'm using 3.10.12, on ubuntu 22.04.

[truher]

closed so i can have the "develop" branch back.

see #1809