Native engine crashes on literal DateTrunc and TimestampTrunc

[andygrove]

Description

When Spark's ConstantFolding optimizer rule is disabled, all-literal trunc() and date_trunc() calls reach the native engine, which crashes with:

• Invalid input to function DateTrunc. Expected (PrimitiveArray<Date32>, Scalar) or (PrimitiveArray<Date32>, StringArray)
• Invalid input to function TimestampTrunc. Expected (PrimitiveArray<TimestampMicrosecondType>, Scalar, String) or (PrimitiveArray<TimestampMicrosecondType>, StringArray, String)

The native functions expect at least one array input but receive all scalars.

How to Reproduce

SET spark.sql.optimizer.excludedRules=org.apache.spark.sql.catalyst.optimizer.ConstantFolding;
SELECT trunc(date('2024-06-15'), 'year');
SELECT date_trunc('year', timestamp('2024-06-15 10:30:45'));

Expected Behavior

The native engine should handle scalar inputs gracefully — either by computing the correct result or by falling back to Spark.

Affected Expressions

• trunc() (DateTrunc)
• date_trunc() (TimestampTrunc)