[BUG] ParseDateTimeSuite UT failed

[pxLi]

Describe the bug
mvn -Dbuildver=302 clean package -DwildcardSuites=com.nvidia.spark.rapids.ParseDateTimeSuite
mvn -Dbuildver=312 clean package -DwildcardSuites=com.nvidia.spark.rapids.ParseDateTimeSuite

ParseDateTimeSuite:
- to_date dd/MM/yy (fall back)
- to_date yyyy-MM-dd *** FAILED ***
  Mismatch between CPU and GPU for the following rows:

  [#40] CPU: [1999-12-29
  ,null]
  [#40] GPU: [1999-12-29
  ,1999-12-29] (ParseDateTimeSuite.scala:302)
- to_date yyyy-MM-dd LEGACY
- to_date yyyy/MM/dd LEGACY
- to_date dd/MM/yyyy
- to_date dd/MM/yyyy LEGACY
- to_date parse date
- to_date parse timestamp
- to_timestamp yyyy-MM-dd *** FAILED ***
  Mismatch between CPU and GPU for the following rows:

  [#35] CPU: [1999-12-29
  ,null]
  [#35] GPU: [1999-12-29
  ,1999-12-29 00:00:00.0] (ParseDateTimeSuite.scala:302)
- to_timestamp dd/MM/yyyy
- to_date default pattern
- unix_timestamp parse date *** FAILED ***
  Mismatch between CPU and GPU for the following rows:

  [#35] CPU: [1999-12-29
  ,null]
  [#35] GPU: [1999-12-29
  ,946425600] (ParseDateTimeSuite.scala:302)

[revans2]

This appears to be for 22.02.0-SNAPSHOT and is not failing on 21.12

[tgravescs]

the failure here is caused when there is a newline in the string:

val ts = Seq("1999-12-29\n")

[tgravescs]

I am not positive on this but this commit is dealing with newline and is about the time our test started to fail:

rapidsai/cudf@c1bfb26
rapidsai/cudf#9620

I might be wrong because its dealing with regex vs string to timestamp but that is the only change I saw in the code. I will verify

[tgravescs]

so I confirmed that reverting 6dc7b1d16f0eecdc2f932a6f9b94af630db534c5 - Fix regex non-multiline EOL/$ matching strings ending with a new-line (#9715)

make the behavior go back and the test passes

[tgravescs]

cudf issue: rapidsai/cudf#9764

[tgravescs]

PR up to temporarily disable the tests, they need to be re-enabled when cudf is fixed

[andygrove]

We should be able to re-enable these tests now.