[BUG] UnsupportedOperationException during delta write with optimize()

[kuhushukla]

Describe the bug
A query that involves a delta table with the optimize call https://docs.databricks.com/en/delta/optimize.html#syntax-examples fails with the Rapids for Spark Plugin, with:

Caused by: java.lang.UnsupportedOperationException: Use getFilePartitions() instead.
at com.databricks.sql.transaction.tahoe.files.TahoeFileIndexWithStaticPartitions.inputFiles(TahoeFileIndexWithStaticPartitions.scala:98)
at org.apache.spark.sql.execution.datasources.HadoopFsRelation.inputFiles(HadoopFsRelation.scala:77)
at com.nvidia.spark.rapids.GpuOverrides.$anonfun$isDeltaLakeMetadataQuery$1(GpuOverrides.scala:4641)
at com.nvidia.spark.rapids.GpuOverrides.$anonfun$isDeltaLakeMetadataQuery$1$adapted(GpuOverrides.scala:4628)

Steps/Code to reproduce bug

import spark.implicits._
// try disabling log query for delta knob, fails either way
spark.conf.set("spark.rapids.sql.detectDeltaLogQueries", "false")
val df0 = sc.parallelize(0 until 10).toDF
val df1 = sc.parallelize(0 until 10).toDF
val df9 = df0.unionAll(df1)

val df00 = sc.parallelize(0 until 100).toDF
val df011 = sc.parallelize(0 until 100).toDF
val df099 = df00.unionAll(df011)

val fin = df9.unionAll(df099)
fin.explain()
sqlContext.sql("CREATE TABLE IF NOT EXISTS test00 (value int) USING delta")
fin.write.mode(SaveMode.Overwrite).format("delta").insertInto("test00")
import io.delta.tables._
val deltaTable = DeltaTable.forName(spark, "test00")
deltaTable.optimize().executeCompaction()

Expected behavior
Non fatal execution.

Environment details (please complete the following information)

• Databricks AWS 13.3, Spark Rapids plugin version 24.06.0,
• 13.3 LTS ML (includes Apache Spark 3.4.1, GPU, Scala 2.12)