[BUG] cache_test failures when testing with 128-bit decimal

[jlowe]

While testing #4823 on Spark 3.2, cache tests were failing when 128-bit decimals were added to decimal_gens. Specifically two tests were failing:

[2022-02-18T19:25:33.459Z] FAILED ../../src/main/python/cache_test.py::test_cache_partial_load[{'spark.sql.inMemoryColumnarStorage.enableVectorizedReader': 'true'}-Decimal(20,2)][ALLOW_NON_GPU(CollectLimitExec)]
[2022-02-18T19:25:33.459Z] FAILED ../../src/main/python/cache_test.py::test_cache_cpu_gpu_mixed[{'spark.sql.inMemoryColumnarStorage.enableVectorizedReader': 'true'}-Decimal(20,2)]

test_cache_partial_load was failing with "not all the plan is on the GPU" while test_cache_cpu_gpu_mixed was failing with:

E                   Caused by: java.lang.ArithmeticException: Decimal precision 21 exceeds max precision 20
E                   	at org.apache.spark.sql.types.Decimal.set(Decimal.scala:122)
E                   	at org.apache.spark.sql.types.Decimal$.apply(Decimal.scala:571)
E                   	at org.apache.spark.sql.types.Decimal.apply(Decimal.scala)
E                   	at org.apache.spark.sql.execution.vectorized.WritableColumnVector.getDecimal(WritableColumnVector.java:378)
E                   	at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage1.processNext(Unknown Source)
E                   	at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
E                   	at org.apache.spark.sql.execution.WholeStageCodegenExec$$anon$1.hasNext(WholeStageCodegenExec.scala:757)
E                   	at org.apache.spark.sql.execution.collect.UnsafeRowBatchUtils$.encodeUnsafeRows(UnsafeRowBatchUtils.scala:80)
E                   	at org.apache.spark.sql.execution.collect.Collector.$anonfun$processFunc$1(Collector.scala:178)
E                   	at org.apache.spark.scheduler.ResultTask.$anonfun$runTask$3(ResultTask.scala:75)
E                   	at com.databricks.spark.util.ExecutorFrameProfiler$.record(ExecutorFrameProfiler.scala:110)
E                   	at org.apache.spark.scheduler.ResultTask.$anonfun$runTask$1(ResultTask.scala:75)
E                   	at com.databricks.spark.util.ExecutorFrameProfiler$.record(ExecutorFrameProfiler.scala:110)
E                   	at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:55)
E                   	at org.apache.spark.scheduler.Task.doRunTask(Task.scala:150)
E                   	at org.apache.spark.scheduler.Task.$anonfun$run$1(Task.scala:119)
E                   	at com.databricks.spark.util.ExecutorFrameProfiler$.record(ExecutorFrameProfiler.scala:110)
E                   	at org.apache.spark.scheduler.Task.run(Task.scala:91)
E                   	at org.apache.spark.executor.Executor$TaskRunner.$anonfun$run$13(Executor.scala:813)
E                   	at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1620)
E                   	at org.apache.spark.executor.Executor$TaskRunner.$anonfun$run$4(Executor.scala:816)
E                   	at scala.runtime.java8.JFunction0$mcV$sp.apply(JFunction0$mcV$sp.java:23)
E                   	at com.databricks.spark.util.ExecutorFrameProfiler$.record(ExecutorFrameProfiler.scala:110)
E                   	at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:672)
E                   	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
E                   	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
E                   	... 1 more