Add Java vectorized scalar function support (1.5)

[staticlibs]

This is a backport of the PR #630 to v1.5-variegata stable branch.

Summary

This PR adds the implementation of Java Scalar Functions (UDFs) in
duckdb-java, using a vectorized callback model for execution.

It introduces function registration, callback bridging, typed vector
read/write APIs, documentation, and test coverage for supported types.

What this PR adds

• New public API on DuckDBConnection:
  • registerScalarFunction(String name, String[] parameterTypes, String returnType, DuckDBVectorizedScalarFunction function)
• New callback contract and vector APIs: - DuckDBVectorizedScalarFunction - DuckDBDataChunkReader - DuckDBReadableVector - DuckDBWritableVector
• JNI/C bridge needed to connect Java callbacks to DuckDB native scalar callback execution
• SQL type parsing helper used by the string-based Java registration API
• Scalar UDF documentation (UDF.MD) and README reference
• Dedicated test suite (TestScalarFunctions) plus binding-level regression tests

Main design decisionsV

1) Prioritize Java-side logic

The design keeps most registration and type wiring logic in Java,
with JNI used only for unavoidable callback bridging
responsibilities.

2) Keep JNI additions minimal and essential

JNI is limited to:

• native callback pointer/state installation
• JVM thread attach/detach from DuckDB execution threads
• callback lifecycle and error propagation
• required helpers for logical type parsing and safe VARCHAR extraction

3) Performance-focused vector path

The UDF execution path uses dedicated typed vector classes
(DuckDBReadableVector/DuckDBWritableVector) instead of generic
JDBC row/object paths, to reduce overhead in callback hot loops:

• primitive typed access/write APIs
• direct output vector writes
• explicit null-mask handling
• reduced boxing/unboxing and object allocation

Correctness and hardening included

• DECIMAL output validates declared precision/scale
• VARCHAR helper validates row bounds
• VARCHAR null rows are guarded in Java and JNI
• Vector code uses ByteOrder.nativeOrder() consistently
• UBIGINT read/write is endian-correct

Testing

• Added broad scalar UDF coverage in TestScalarFunctions