stats_glm.t fails with PDL 2.086

[sebastic]

The test suite for the Debian package fails since the update of pdl to 2.086:

 34s Error in PDL::Slatec::gedi: output parameter 'det' implicit dim 1 size 2, can't broadcast over output ndarray with size > 1
 34s  at slatec.pd line 463.
 34s 	PDL::matinv(PDL=SCALAR(0x55aeb5a5fcc0)) called at glm.pd line 766
 34s 	PDL::Stats::GLM::_ols_common(1, PDL=SCALAR(0x55aeb5a5fc30), PDL=SCALAR(0x55aeb5a5fa98)) called at glm.pd line 727
 34s 	PDL::ols_t(PDL=SCALAR(0x55aeb5a5fc30), PDL=SCALAR(0x55aeb5a5fa98)) called at t/stats_glm.t line 54
 34s # Tests were run but no plan was declared and done_testing() was not seen.
 34s # Looks like your test exited with 2 just after 11.
 34s Dubious, test returned 2 (wstat 512, 0x200)

https://ci.debian.net/packages/libp/libpdl-stats-perl/testing/amd64/44766893/

CC: @mohawk2

[mohawk2]

Thank you. The downstream testing didn't catch that, and I guess that's because it doesn't make sure Fortran is there so Slatec doesn't get built. That's clearly an error in PDL-Stats which I will fix right now. (I'm also adding 32-bit CI to avoid repeats of PDLPorters/pdl#469)

[mohawk2]

The above-linked commit catches this problem. Now to fix it!

[mohawk2]

@sebastic The above commit fixes this, and an updated PDL will be released soon. Are you happy to close this, since it's not actually a bug in PDL::Stats?

[sebastic]

Sure, I can reopen it if the issue is still present after the update of the pdl package.