[spark] Add max_pt function

[ulysses-you]

Purpose

Adds a new Spark function max_pt. It accpets a string type literal and return a max-valid-toplevel partition value.

• valid means the partition contains data files
• toplevel means only return the first partition value if the table has multi-partition columns

max_pt will throw exception when:

• the table is not a partitioned table
• the partitioned table does not have partition
• all of the partitions do not contains data files

This pr adds max_pt through spark v2 function catalog using a fake scalar function and then adds a rule to replace max_pt to a literal value during analysis.

Example:

SELECT max_pt('t')
=>
20250101

API and Format

no

Documentation

add sql-functions docs

[JingsongLi]

Maybe add a new documentation: sql-function?

[ulysses-you]

@JingsongLi added docs

[JingsongLi]

+1