[feature]: Add Spark 4.0 VARIANT type and Parquet read support

[oarap]

Summary

Implements end-to-end support for the Spark 4.0 VARIANT type in Bolt — from the type system and vector representation through Parquet I/O, serialization, expression evaluation, and Arrow interop.

The VARIANT type stores semi-structured JSON-like data as two VARBINARY fields: value (binary payload) and metadata (string dictionary).

What problem does this PR solve?

Issue Number: close #420

Type of Change

• 🐛 Bug fix (non-breaking change which fixes an issue)
• ✨ New feature (non-breaking change which adds functionality)
• 🚀 Performance improvement (optimization)
• ⚠️ Breaking change (fix or feature that would cause existing functionality to change)
• 🔨 Refactoring (no logic changes)
• 🔧 Build/CI or Infrastructure changes
• 📝 Documentation only

Description

Changes

Type System

• Added TypeKind::VARIANT, VariantType, VariantValue (view) and OwnedVariantValue (owning) with full folly::hasher / std::hash support.
• VARIANT is marked non-orderable and non-comparable — byte-level comparison does not imply semantic equality, so GROUP BY, JOIN, and ORDER BY on VARIANT are blocked at the planner level.

Vector Layer

• New VariantVector (VectorEncoding::Simple::VARIANT): a BaseVector subclass backed by two child FlatVector<StringView> columns for value and metadata.
• Integrated into DecodedVector, BaseVector::createInternal, VectorEncoding, and related utilities.

Parquet Reader

• New VariantColumnReader (subclass of StructColumnReader) promotes STRUCT<value BINARY, metadata BINARY> Parquet columns into VariantVector outputs.
• Auto-detection heuristic in getParquetColumnInfo; explicit requestedType = VARIANT always takes precedence. DWRF footer parsing deliberately does not auto-promote to avoid false positives on legitimate structs.

Spark VARIANT Encoding/Decoding

• SparkVariantEncoder: two-pass JSON → VARIANT binary encoder.
• SparkVariantReader: decoder supporting Spark bit-coded, compact, and raw JSON formats with auto-detection. decode() returns std::optional<std::string> — std::nullopt = failure, "null" = valid JSON null.
• JSONPath navigation (.key, [index], quoted keys) on binary payloads without a full decode.

SQL Functions

• parse_json(VARCHAR) → VARIANT: encodes a JSON string into Spark binary VARIANT.
• variant_get(VARIANT, VARCHAR) → VARCHAR: extracts a value at a JSONPath via three strategies: binary nav → compact nav → full decode + simdjson fallback.

Tests

• VariantTypeTest, VariantVectorTest, VariantFunctionTest — type, vector, and function correctness.
• 5 new ParquetReaderTest cases covering end-to-end reads, ScanSpec order mismatches, all Spark primitive types, nested containers (ARRAY<VARIANT>, MAP<VARCHAR, VARIANT>, STRUCT<VARIANT, VARIANT>), and raw JSON payloads. Real Parquet fixture files included.

Performance Impact

• No Impact: This change does not affect the critical path (e.g., build system, doc, error handling).

• Positive Impact: I have run benchmarks.

  Click to view Benchmark Results

  Paste your google-benchmark or TPC-H results here.
  Before: 10.5s
  After:   8.2s  (+20%)
  

• Negative Impact: Explained below (e.g., trade-off for correctness).

Release Note

Please describe the changes in this PR

Release Note:

Release Note:
- Read `VARIANT` columns from Parquet files written by Spark 4.0. Bolt automatically detects the physical `STRUCT<value BINARY, metadata BINARY>` encoding and promotes it to a first-class `VARIANT` type.

- New SQL functions:
  - `parse_json(VARCHAR) → VARIANT` — parses a JSON string into Spark binary VARIANT format.
  - `variant_get(VARIANT, VARCHAR) → VARCHAR` — extracts a value from a VARIANT column using a JSONPath expression (e.g. `$.a.b`, `$[0]`).

Checklist (For Author)

• I have added/updated unit tests (ctest).
• I have verified the code with local build (Release/Debug).
• I have run clang-format / linters.
• (Optional) I have run Sanitizers (ASAN/TSAN) locally for complex C++ changes.
• No need to test or manual test.

Breaking Changes

• No

• Yes (Description: ...)

  Click to view Breaking Changes

  Breaking Changes:
  - Description of the breaking change.
  - Possible solutions or workarounds.
  - Any other relevant information.