debug inits tests

[jgabry]

Don't merge this. Just additional tests related to #937 (comment) as discussed with @SteveBronder

[SteveBronder]

It looks like it failed because of something else?

[jgabry]

It's just failing on R 3.6 because we're now requiring R 4.0. Just merged a PR into this one that should fix that. Aside from the check using R 3.6 I don't see any failures yet.

[jgabry]

Ugh, same error. Really strange. Not sure what's going on.

https://github.com/stan-dev/cmdstanr/actions/runs/8790487616/job/24122605861#step:13:3310

── Failure ('test-fit-init-debug.R:61:3'): Pathfinder method with psis_resample as false works as init ──
Expected `test_inits(mod_logistic, fit_path_init, data_list_logistic)` to run without any errors.
ℹ Actually got a <simpleError> with text:
  Assertion on 'files' failed: File does not exist: '/tmp/RtmpNMWjuB/working_dir/RtmpIOmJZ0/logistic-202404222153-1-84d8fc.csv'.

[jgabry]

@SteveBronder now it's passing everything here, but not sure if that's just because we got lucky (since the error seems nondeterministic). I will trigger the tests to run again.

[SteveBronder]

@jgabry I think I figured it out but it's very confusing. Removing the set.seed() in the file I changed above will allow let the while loop run forever instead of failing to generate the file. I don't know about about R internals to understand why that would be the issue though? but I've had it that while loop for 10 minutes now without issue

[andrjohns]

Closing as issue resolved