Disable test under MIRI which appears to exceed memory limits intermittently on github CI

[alamb]

Which issue does this PR close?

Closes #879

Rationale for this change

The MIRI CI check is failing intermittently on master. The symptoms of the failing MIRI tests seem to be the same as would happen if miri is killed by the github OOM killer -- see #879 (comment)

We can't reproduce any MIRI failures locally so this seems like either a bug in MIRI or it is using too many resources for github.

Note the test in question was added relatively recently in #706 by @b41sh (but then MIRI started failing after updating to a new rust version)

What changes are included in this PR?

Disable the failing test

Are there any user-facing changes?

No

[alamb]

Bah, now it blows up somewhere else later: https://github.com/apache/arrow-rs/runs/4097468828?check_suite_focus=true

[codecov-commenter]

Codecov Report

Merging #910 (41b4f3b) into master (1d3d5e3) will increase coverage by 0.00%.
The diff coverage is 100.00%.

❗ Current head 41b4f3b differs from pull request most recent head a85e6cd. Consider uploading reports for the commit a85e6cd to get more accurate results

@@           Coverage Diff           @@
##           master     #910   +/-   ##
=======================================
  Coverage   82.29%   82.30%           
=======================================
  Files         168      168           
  Lines       48028    48028           
=======================================
+ Hits        39527    39529    +2     
+ Misses       8501     8499    -2     

Impacted Files	Coverage Δ
arrow/src/compute/kernels/comparison.rs	92.63% <100.00%> (ø)
arrow/src/datatypes/datatype.rs	64.93% <0.00%> (-0.44%)	⬇️
arrow/src/array/transform/mod.rs	85.47% <0.00%> (+0.13%)	⬆️
parquet/src/encodings/encoding.rs	93.71% <0.00%> (+0.19%)	⬆️
parquet_derive/src/parquet_field.rs	66.21% <0.00%> (+0.22%)	⬆️

---

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 1d3d5e3...a85e6cd. Read the comment docs.

[houqp]

Have we tried different Rust versions?

[jimexist]

Have we tried different Rust versions?

yes i think by default this runs on latest nightly

[alamb]

Yeah, I tried latest nightly in #907 which did not seem to solve the issue either.

[alamb]

I am considering adding a loop that retries running MIRI a few times to see if we can get a successful run. Not the most elegant of solutions, but one that has worked for me it the past. New PR shortly

[jimexist]

I am considering adding a loop that retries running MIRI a few times to see if we can get a successful run. Not the most elegant of solutions, but one that has worked for me it the past. New PR shortly

How about sharding the Miri tests so that no container runs too many tests

[alamb]

How about sharding the Miri tests so that no container runs too many tests

Good idea. I will try that if #922 doesn't pan out

#922 has a retry loop