[GLUTEN-6625] [VL] Handle Array with null value

[vectorijk]

What changes are proposed in this pull request?

Fixes #6625

How was this patch tested?

Was this patch authored or co-authored using generative AI tooling?

[github-actions]

Run Gluten Clickhouse CI on x86

[vectorijk]

Run Velox Backend (ARM)

[vectorijk]

Run Gluten ClickHouse CI

[vectorijk]

/ansi-test

[github-actions]

🔄 ANSI mode analysis started by @vectorijk. View run

[github-actions]

ANSI Mode Test Analysis Report (Spark 4.1)

Note

Expression-level ANSI mode offload coverage analysis.
Test config: spark.sql.ansi.enabled=true, spark.gluten.sql.ansiFallback.enabled=false.

• Passed (🟢): Velox correctly handles ANSI semantics
• Fallback (🔴): Expression falls back to Spark execution, needs ANSI support in Velox
• Failed (🟡): Velox executes but ANSI error behavior differs from Spark, needs exception handling fix

ANSI Offload suites: 76 tests, 11686 records | Other suites: 3442 tests

ANSI Offload

Overview (ANSI Offload Expression Records)

Classification	Count	%
🟢 Passed	11178	95.7%
🟡 Failed	11	0.1%
🔴 Fallback	497	4.3%

Per-Suite Summary

Suite	🟢 Passed	🟡 Failed	🔴 Fallback
GlutenTryCastSuite	11160 (96%)	10	497
GlutenDecimalExpressionSuite	18 (95%)	1	0

Failure Cause Analysis (11 failures)

Cause	Count	Description
WRONG_EXCEPTION	1	Exception wrapped as SparkException
OTHER	10	Result mismatch or eval exception

Other (63 failures)

Suite	Failures
ScalarFunctionsValidateSuite	11
GlutenHiveTableScanSuite	10
GlutenDataFrameFunctionsSuite	7
MiscOperatorSuite	5
DateFunctionsValidateSuite	4
MathFunctionsValidateSuite	4
FallbackSuite	fallback on spilt with unsupported regex fallback when nested loop join has unsupported expression fallback when join post filter has unsupported expression
GlutenBloomFilterAggregateQuerySuiteCGOff	Test NULL inputs for might_contain SPARK-54336: Fix BloomFilterMightContain type check with ScalarSubqueryReference Gluten - Test that might_contain on bloom_filter_agg with empty input
GlutenParquetV2PartitionDiscoverySuite	Various partition value types Various inferred partition value types SPARK-40212: SparkSQL castPartValue does not properly handle byte, short, float
VeloxScanSuite	parquet index based schema evolution ORC positional schema evolution
GlutenComplexTypeSuite	SPARK-33386: GetArrayItem ArrayIndexOutOfBoundsException CreateMap
VeloxLiteralSuite	Literal Fallback
VeloxOrcDataTypeValidationSuite	Decimal type
VeloxParquetDataTypeValidationSuite	Decimal type
VeloxStringFunctionsSuite	regex invalid
VeloxTPCHDistinctSpillSuite	distinct spill
UDFPartialProjectSuite	udf in agg simple
GlutenCSVLegacyTimeParserSuite	SPARK-36831: Support reading and writing ANSI intervals
GlutenCSVv1Suite	SPARK-36831: Support reading and writing ANSI intervals
GlutenPartitionedWriteSuite	timeZone setting in dynamic partition writes

[github-actions]

Run Gluten Clickhouse CI on x86

[vectorijk]

/ansi-test

[github-actions]

🔄 ANSI mode analysis started by @vectorijk. View run

[github-actions]

ANSI Mode Test Analysis Report (Spark 4.1)

Note

Expression-level ANSI mode offload coverage analysis.
Test config: spark.sql.ansi.enabled=true, spark.gluten.sql.ansiFallback.enabled=false.

• Passed (🟢): Velox correctly handles ANSI semantics
• Fallback (🔴): Expression falls back to Spark execution, needs ANSI support in Velox
• Failed (🟡): Velox executes but ANSI error behavior differs from Spark, needs exception handling fix

ANSI Offload suites: 76 tests, 11779 records | Other suites: 3442 tests

ANSI Offload

Overview (ANSI Offload Expression Records)

Classification	Count	%
🟢 Passed	11167	94.8%
🟡 Failed	1	0.0%
🔴 Fallback	611	5.2%

Per-Suite Summary

Suite	🟢 Passed	🟡 Failed	🔴 Fallback
GlutenTryCastSuite	11149 (95%)	0	611
GlutenDecimalExpressionSuite	18 (95%)	1	0

Failure Cause Analysis (1 failures)

Cause	Count	Description
WRONG_EXCEPTION	1	Exception wrapped as SparkException

Other (40 failures)

Suite	Failures
ScalarFunctionsValidateSuite	10
GlutenHiveTableScanSuite	10
MiscOperatorSuite	5
MathFunctionsValidateSuite	4
VeloxScanSuite	parquet index based schema evolution ORC positional schema evolution
DateFunctionsValidateSuite	make_date read as timestamp_ntz
FallbackSuite	fallback when nested loop join has unsupported expression
VeloxLiteralSuite	Literal Fallback
VeloxOrcDataTypeValidationSuite	Decimal type
VeloxParquetDataTypeValidationSuite	Decimal type
UDFPartialProjectSuite	udf in agg simple
GlutenDataFrameFunctionsSuite	Gluten - map with arrays
GlutenComplexTypeSuite	SPARK-33386: GetArrayItem ArrayIndexOutOfBoundsException