[WIP][SPARK-51423][SQL] Add the current_time() function for TIME datatype

[the-sakthi]

What changes were proposed in this pull request?

This PR adds support for a new function current_time() which returns the current time at the start of query evaluation.

scala> spark.sql("SELECT current_time(0);").show()
+--------------+
|current_time()|
+--------------+
|      04:48:47|
+--------------+


scala> spark.sql("SELECT current_time();").show()
+---------------+
| current_time()|
+---------------+
|04:48:51.609168|
+---------------+


scala> spark.sql("SELECT current_time(3);").show()
+--------------+
|current_time()|
+--------------+
|  04:48:54.846|
+--------------+


scala> spark.sql("SELECT current_time(6);").show()
+---------------+
| current_time()|
+---------------+
|04:48:59.222348|
+---------------+

scala> spark.sql("SELECT current_time(7);").show()
org.apache.spark.sql.catalyst.ExtendedAnalysisException: [DATATYPE_MISMATCH.TYPE_CHECK_FAILURE_WITH_HINT] Cannot resolve "current_time()" due to data type mismatch: Invalid precision 7. Must be between 0 and 6.. SQLSTATE: 42K09; line 1 pos 7;
'Project [unresolvedalias(current_time(7))]
+- OneRowRelation

  at org.apache.spark.sql.catalyst.analysis.package$AnalysisErrorAt.failAnalysis(package.scala:52)
  ...

Why are the changes needed?

Adds a built-in current_time([n]) function returning just the time portion (in a TIME(n) type). This aligns Spark with other SQL systems offering a native time function, improves convenience for time-only queries, and complements existing functions like current_date and current_timestamp.

Does this PR introduce any user-facing change?

Yes, adds a new function. Users can now get the current time using this function.

How was this patch tested?

Manual testing.

Pending

• Support for current_time [i.e. without any braces]
• Implementation of codegen (currently extending CodegenFallback)
• Addition of UTs.
• Use inbuilt functions for computeMicrosAtPrecision

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

No

[MaxGekk]

... which returns the current time at the start of query evaluation.

I think the current implementation doesn't satisfy the requirement, and every call in a query returns new current time. Could you look at the CurrentTimestampLike, and the rule:

spark/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/finishAnalysis.scala

Lines 108 to 111 in 2822526

	/**
	* Computes the current date and time to make sure we return the same result in a single query.
	*/
	object ComputeCurrentTime extends Rule[LogicalPlan] {

[the-sakthi]

Ahh sure! Thanks for the early review, Max. Will refer CurrentTimestampLike to better cater to the requirement.