perf: optimize native shuffle struct field processing with field-major order #4311

Job	Run time
macos-14/Spark 3.4, JDK 11, Scala 2.12 [exec]	52m 14s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [exec]	51m 22s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [fuzz]	1h 18m 9s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [fuzz]	1h 25m 28s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [parquet]	1h 1m 4s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [sql]	44m 2s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [shuffle]	1h 5m 54s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [expressions]	56m 50s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [exec]	49m 53s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [sql]	30m 11s
macos-14/Spark 3.4, JDK 11, Scala 2.12 [fuzz]	1h 40m 14s
macos-14/Spark 4.0, JDK 17, Scala 2.13 [parquet]	52m 27s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [expressions]	57m 26s
macos-14/Spark 3.4, JDK 11, Scala 2.12 [sql]	37m 13s
macos-14/Spark 3.5, JDK 17, Scala 2.13 [shuffle]	1h 1m 16s
macos-14/Spark 3.4, JDK 11, Scala 2.12 [expressions]	58m 22s
macos-14/Spark 3.4, JDK 11, Scala 2.12 [shuffle]	1h 15m 49s
macos-14/Spark 3.4, JDK 11, Scala 2.12 [parquet]	59m 16s
	17h 57m 10s

Provide feedback