feat: add register table for catalogs

[Elbehery]

1. PositionDeleteFileWriter (crates/iceberg/src/writer/base_writer/position_delete_writer.rs)

• Schema Implementation: Fixed schema with file_path (String) and pos (Long) columns
• Field IDs: Correct Iceberg spec field IDs (2147483546, 2147483545)
• Writer Integration: Full IcebergWriter trait implementation
• Test Coverage: Unit tests for schema validation, batch creation, and writer functionality
• Module Integration: Properly exposed in mod.rs

2. Equality Delete Parsing (crates/iceberg/src/arrow/caching_delete_file_loader.rs)

• RecordBatch Processing: Converts Arrow batches to Iceberg predicates
• Multi-type Support: Handles all Arrow data types (String, Int, Float, Date, etc.)
• Logical Operations: Constructs proper AND/OR predicate trees
• Error Handling: Robust handling of missing field IDs and null values

3. LZ4 Compression Support (crates/iceberg/src/puffin/)

• Dependency Added: lz4_flex = "0.11" in Cargo.toml
• Compression Logic: lz4_flex::compress_prepend_size() implementation
• Decompression Logic: lz4_flex::decompress_size_prepended() implementation
• Test Updates: All LZ4 tests now expect success instead of FeatureUnsupported

4. Delete Manifest Processing (crates/iceberg/src/transaction/snapshot.rs)

• SnapshotProducer Extensions: Added added_delete_files field
• Validation Logic: add_delete_files() with content type validation
• Manifest Generation: write_delete_manifest() for delete entries
• Summary Integration: Delete files included in snapshot summaries

5. Partition Support (crates/iceberg/src/writer/file_writer/parquet_writer.rs)

• Hive-style Parsing: Extracts col=value patterns from file paths
• Type Conversion: Converts string values to appropriate Literal types
• Base64 Support: Handles binary partition values
• Integration: Fully integrated into parquet_files_to_data_files()

6. NaN Value Counts (crates/iceberg/src/writer/file_writer/parquet_writer.rs)

• Statistics Extraction: Reads Parquet row group statistics
• NaN Detection: Uses NanValueCountVisitor for float/double columns
• Async Implementation: Proper async file I/O handling
• Integration: Passed to DataFileBuilder construction

7. Deletion Vector Support (crates/iceberg/src/arrow/caching_delete_file_loader.rs)

• Puffin Detection: is_puffin_file() helper function
• Blob Parsing: parse_delete_vector_blob() with RoaringTreemap
• Index Integration: Added to PopulatedDeleteFileIndex
• Context Enum: New DelVecs variant in DeleteFileContext

8. Case-Insensitive Predicate Binding (crates/iceberg/src/scan/)

• Context Propagation: Added case_sensitive to FileScanTask, ManifestEntryContext, ManifestFileContext
• Filter Integration: Used in build_equality_delete_predicate()
• End-to-End Flow: Properly passed through scan planning pipeline

Fixes #1518

[liurenjie1024]

Hi, @Elbehery Is this pr really for resigtering table for sql catalog? I see most changes are unrelated.

[Elbehery]

Hi, @Elbehery Is this pr really for resigtering table for sql catalog? I see most changes are unrelated.

oh deep apologies, would you kindly review or correct my path ?

i tried as much as i can, i am very new to Icerberg :/

[liurenjie1024]

These two prs are implementing register_table for memory and rest catalog, you could take them as a reference:
#1549
#1521

[Elbehery]

These two prs are implementing register_table for memory and rest catalog, you could take them as a reference: #1549 #1521

thanks so much 🙏🏽