[REVIEW] Upgrade numba pinning to be in-line with rest of rapids

[galipremsagar]

This PR upgrades minimum version pinning of numba to 0.54 from 0.53.1 to be in-sync with rest of the rapids packages:
https://github.com/rapidsai/integration/blob/branch-22.06/conda/recipes/versions.yaml#L114

[pentschev]

LGTM, thanks @galipremsagar !

[codecov-commenter]

Codecov Report

Merging #912 (2ae28cb) into branch-22.06 (497180d) will not change coverage.
The diff coverage is n/a.

@@              Coverage Diff              @@
##           branch-22.06    #912    +/-   ##
=============================================
  Coverage          0.00%   0.00%            
=============================================
  Files                22      16     -6     
  Lines              3075    2094   -981     
=============================================
+ Misses             3075    2094   -981     

Impacted Files	Coverage Δ
dask_cuda/benchmarks/local_cudf_merge.py
dask_cuda/benchmarks/utils.py
dask_cuda/benchmarks/local_cudf_shuffle.py
dask_cuda/benchmarks/local_cupy.py
dask_cuda/_version.py
dask_cuda/benchmarks/local_cupy_map_overlap.py

---

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 497180d...2ae28cb. Read the comment docs.

[galipremsagar]

rerun tests

[galipremsagar]

@pentschev does the CI failure(a pytest timeout) look specific to this PR?

[pentschev]

I'm not quite sure to be honest, I actually have seen it fail in a few occasions. @madsbk do you think we need to increase the timeout for test_worker_force_spill_to_disk or is it flaky for some reason?

[madsbk]

I'm not quite sure to be honest, I actually have seen it fail in a few occasions. @madsbk do you think we need to increase the timeout for test_worker_force_spill_to_disk or is it flaky for some reason?

Let's try to increase the timeout, if it keep failing we can also skip it.
I am working on a complete new spilling approach, which hopefully makes all of this obsolete :)

[galipremsagar]

Looks like increasing timeout worked, @pentschev | @madsbk could any one of you review the new changes?

[pentschev]

Thanks @galipremsagar and @madsbk !

[pentschev]

@gpucibot merge