[REVIEW] Change pytest distribution algorithm and increase parallelism in CI

[galipremsagar]

This PR increases the number of cores allocated to pytest-xdist and also changes distribution algorithm for cudf to loadscope.

This PR:

custreamz:
= 600 passed, 6 skipped, 145 xfailed, 96 xpassed, 7 warnings in 88.10s (0:01:28) =

dask-cudf:
= 891 passed, 13 skipped, 2 xfailed, 1 xpassed, 115 warnings in 63.42s (0:01:03) =

cudf(loadscope):
= 88743 passed, 2408 skipped, 985 xfailed, 1978 xpassed, 3115 warnings in 1195.87s (0:19:55) =

branch-22.04:

custreamz:

= 600 passed, 6 skipped, 145 xfailed, 96 xpassed, 5 warnings in 102.16s (0:01:42) =

dask-cudf:

= 891 passed, 13 skipped, 2 xfailed, 1 xpassed, 113 warnings in 78.00s (0:01:18) =

cudf:

= 88745 passed, 2408 skipped, 985 xfailed, 1978 xpassed, 3113 warnings in 2249.52s (0:37:29) =

[codecov]

Codecov Report

Merging #10232 (ebb6df8) into branch-22.04 (a7d88cd) will increase coverage by 0.04%.
The diff coverage is 0.00%.

@@               Coverage Diff                @@
##           branch-22.04   #10232      +/-   ##
================================================
+ Coverage         10.42%   10.47%   +0.04%     
================================================
  Files               119      122       +3     
  Lines             20603    20502     -101     
================================================
- Hits               2148     2147       -1     
+ Misses            18455    18355     -100     

Impacted Files	Coverage Δ
python/cudf/cudf/_fuzz_testing/fuzzer.py	0.00% <0.00%> (ø)
python/cudf/cudf/_fuzz_testing/io.py	0.00% <0.00%> (ø)
python/cudf/cudf/_fuzz_testing/main.py	0.00% <0.00%> (ø)
python/cudf/cudf/_version.py	0.00% <0.00%> (ø)
python/cudf/cudf/comm/gpuarrow.py	0.00% <0.00%> (ø)
python/cudf/cudf/core/_base_index.py	0.00% <0.00%> (ø)
python/cudf/cudf/core/column/categorical.py	0.00% <0.00%> (ø)
python/cudf/cudf/core/column/column.py	0.00% <0.00%> (ø)
python/cudf/cudf/core/column/datetime.py	0.00% <ø> (ø)
python/cudf/cudf/core/column/numerical.py	0.00% <0.00%> (ø)
... and 45 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 2e458b9...ebb6df8. Read the comment docs.

[bdice]

Discussed in Slack thread -- looks good to me, and appears to cut the runtime by 45%. As long as 8 cores doesn't thrash the machine in a way that negatively affects other CI things (e.g. logging? background processes?) or lead to GPU out-of-memory errors, this looks good to me.

[galipremsagar]

@gpucibot merge