[lake/paimon] Support NestedRow types for tiering paimon

[XuQianJin-Stars]

Purpose

Linked issue: close #2251

This PR adds support for NestedRow types in Paimon lake tiering.

Brief change log

• Modified FlussRowAsPaimonRow.getRow() to support nested row fields by creating new FlussRowAsPaimonRow instances recursively
• Modified FlussArrayAsPaimonArray.getRow() to support arrays containing nested row elements
• Added comprehensive test cases in FlussRecordAsPaimonRowTest covering various nested row scenarios:
  • Simple nested rows with primitive types
  • Deeply nested rows (row within row)
  • Arrays of nested rows
  • Nested rows containing array fields
  • Nested rows with all primitive types (boolean, byte, short, int, long, float, double, string)
  • Null nested rows and nullable fields within nested rows
  • Nested rows with complex types (Decimal, Timestamp)

Tests

Unit Tests:

• FlussRecordAsPaimonRowTest#testSimpleNestedRow - Tests basic nested row conversion
• FlussRecordAsPaimonRowTest#testDeeplyNestedRow - Tests multi-level nested rows
• FlussRecordAsPaimonRowTest#testArrayOfNestedRows - Tests arrays containing nested rows
• FlussRecordAsPaimonRowTest#testNestedRowWithArrayField - Tests nested rows with array fields
• FlussRecordAsPaimonRowTest#testNestedRowWithAllPrimitiveTypes - Tests all primitive types in nested rows
• FlussRecordAsPaimonRowTest#testNullNestedRow - Tests null nested row handling
• FlussRecordAsPaimonRowTest#testNestedRowWithNullableFields - Tests nullable fields in nested rows
• FlussRecordAsPaimonRowTest#testNestedRowWithDecimalAndTimestamp - Tests complex types in nested rows

All tests verify correct data conversion between Fluss nested rows and Paimon nested rows.

API and Format

API: No public API changes

Format: No storage format changes. This change only affects the internal data conversion layer between Fluss and Paimon.

Documentation

New Feature: No

This change extends existing Paimon lake tiering functionality to support NestedRow types, which is an internal enhancement. The data type conversion mapping table in the documentation (website/docs/streaming-lakehouse/integrate-data-lakes/paimon.md) already covers ROW types, so no documentation update is required.

[Copilot]

Pull request overview

This PR adds support for NestedRow types in Paimon lake tiering by implementing the previously unsupported getRow() methods in the data conversion layer between Fluss and Paimon.

Key Changes:

• Implemented recursive nested row conversion by creating new FlussRowAsPaimonRow instances for nested row fields
• Extended array support to handle arrays containing nested row elements
• Added 8 comprehensive test cases covering simple nested rows, deeply nested structures, arrays of rows, null handling, and various data types

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
FlussRowAsPaimonRow.java	Implemented getRow() method to convert nested Fluss rows to Paimon rows recursively with proper null handling
FlussArrayAsPaimonArray.java	Implemented getRow() method to support arrays containing nested row elements
FlussRecordAsPaimonRowTest.java	Added 8 test methods covering nested row scenarios including primitive types, complex types, null handling, deep nesting, and arrays of rows

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[luoyuxia]

@XuQianJin-Stars Thanks for the pr. Let's add two test:

• add test in PaimonRowAsFlussRow for nestrow type
• add IT for nested-row type, you can refer to what we do for array type

[luoyuxia]

Could you please combine all test methods for NestedRow to a single method just like main branch for array type?
Two many methods make it hard to track...

[XuQianJin-Stars]

@luoyuxia Hi, i already updated the pr. Please help review when you got some time

[wuchong]

Replace by @ValueSource(booleans = {true, false}) and remove tieringNestedRowWriteArgs method.

[XuQianJin-Stars]

@wuchong Hi, i already updated the pr. Please help review when you got some time.