Revert "Support barrier mode for mapInPandas/mapInArrow (#10343)"

[wbo4958]

Fixes #10359.

This reverts commit 6723a68, which seems to have caused a compile error on Databricks:

01:49:20  [ERROR] /home/ubuntu/spark-rapids/sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuOverrides.scala:70: object GpuMapInPandasExecMeta is not a member of package org.apache.spark.sql.rapids.shims
01:49:20  [ERROR] import org.apache.spark.sql.rapids.shims.{GpuAscii, GpuMapInPandasExecMeta, GpuTimeAdd}
01:49:20  [ERROR]        ^
01:49:22  [ERROR] /home/ubuntu/spark-rapids/sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuOverrides.scala:4356: not found: type GpuMapInPandasExecMeta
01:49:22  [ERROR]       (mapPy, conf, p, r) => new GpuMapInPandasExecMeta(mapPy, conf, p, r)),

[jlowe]

build

[jlowe]

@wbo4958 It would be nice to have a blurb in the PR description explaining why we're doing this, otherwise team members have to dig to try to figure out why we would want this and what issue it solves to do so.

[jlowe]

build

[mythrocks]

Pardon me, I've taken the liberty to add some context to the description.

[wbo4958]

Hi @jlowe, I got the info from @GaryShen2008 that 24.02 branch was burned down, so I reverted this PR in 24.02 and will merge it in 24.04. Sorry for the trouble.

[sameerz]

@wbo4958 do we need to reopen the original issue #10344 and retarget it to 24.04?

[wbo4958]

@wbo4958 do we need to reopen the original issue #10344 and retarget it to 24.04?

Yeah, I've just opened it.