Remove xfail for KBinsDiscretizer quantile tests

[wphicks]

Following the update to cupy 8.5.0, the bad read in the cupy.percentile kernel should no longer be an issue, allowing us to remove the xfail on this test.

Close #2933

[dantegd]

@wphicks I know that this might be overkill, but do you think it'd be a good idea to run the test a large number of times locally, though a CI run that only runs that test a number of times might be good too, be a good idea?

[codecov-commenter]

Codecov Report

❗ No coverage uploaded for pull request base (branch-0.20@f87ca91). Click here to learn what that means.
The diff coverage is n/a.

@@              Coverage Diff               @@
##             branch-0.20    #3804   +/-   ##
==============================================
  Coverage               ?   86.04%           
==============================================
  Files                  ?      225           
  Lines                  ?    17113           
  Branches               ?        0           
==============================================
  Hits                   ?    14725           
  Misses                 ?     2388           
  Partials               ?        0           

Flag	Coverage Δ
dask	49.29% <0.00%> (?)
non-dask	77.95% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f87ca91...c3fdc55. Read the comment docs.

[wphicks]

do you think it'd be a good idea to run the test a large number of times locally, though a CI run that only runs that test a number of times might be good too, be a good idea?

I don't mind doing that again if it would give us more confidence in this, but as I was initially investigating #3481, I ran this some huge number of times (tens of thousands maybe?) without seeing a failure. I also have a fair amount of confidence that we found the root cause on this one, though it's always possible that I'm wrong. Want me to spin up another run overnight, or do we feel good enough with that level of testing?

[dantegd]

I think that level of testing is good, I had forgotten the journey of the bug a little bit, so will go ahead and merge

[dantegd]

@gpucibot merge