[Kernel][Expression] - Performance Optimization for LIKE expression evaluation

Bug

Which Delta project/connector is this regarding?

[ ] Spark
[ ] Standalone
[ ] Flink
[x] Kernel
[ ] Other (fill in here)

Describe the problem

LIKE expression implementation can take advantage of a performance optimization to perform java regex conversion only once when the pattern input is a a literal. In the current implementation java regex conversion happens for every row in the input.

Steps to reproduce

for e.g. for the expression col("c1") LIKE 'a%' will invoke LikeExpressionEvaluator.escapeLikeRegex(...) for every single row in the column vector for c1.

Observed results

LikeExpressionEvaluator.escapeLikeRegex(...) for every single row in the column vector expression is evaluated against even when pattern is static.

Expected results

LikeExpressionEvaluator.escapeLikeRegex(...) is invoked only once for the entire column vector when pattern is static.

Environment information

Delta Lake version: 3.3.0
Spark version: N/A
Scala version: 2.12.x & 2.13.x

Willingness to contribute

The Delta Lake Community encourages bug fix contributions. Would you or another member of your organization be willing to contribute a fix for this bug to the Delta Lake code base?

[x] Yes. I can contribute a fix for this bug independently.
[ ] Yes. I would be willing to contribute a fix for this bug with guidance from the Delta Lake community.
[ ] No. I cannot contribute a bug fix at this time.

delta-io / delta