Spark 3.1: fill ORC initial defaults on the row reader

[cbb330]

Summary

• Wire the Spark 3.1 row ORC reader to materialize a field's initial-default when it is absent from an id-bearing file, reusing BaseDataReader's constant converter.
• Mirrors the Generic reader behavior from ORC: fill initial defaults for missing id-bound fields #261 for the Spark 3.1 row path.

Stack (bottom -> top)

1. ORC: bind struct fields by Iceberg field id in generic and Spark 3.1 readers #256 - ORC id-binding backport (base 1.2.x)
2. API: add column-default model support #260 - API column-default model (base chbush/orc-id-binding-backport)
3. ORC: fill initial defaults for missing id-bound fields #261 - ORC initial-default fill (base chbush/orc-column-default-api)
4. this PR - Spark 3.1 wiring (base chbush/orc-initial-default-orc-fill)

Depends on #261 (and transitively #260, #256). Review this PR's own delta against its base branch chbush/orc-initial-default-orc-fill.

Supported scope

• Spark 3.1 ORC row reader

Vectorized ORC, other Spark/Flink readers, and predicate pushdown on omitted default columns remain deferred.

Testing Done

• Spark 3.1 row tests: declared, undeclared, and nested defaults
• Full Spark 3.1 test suite passes
• Spark 3.1 checkstyle and formatting pass

---

Tests generated with unit-tests plugin

Made with Cursor

[cbb330]

Closing: this was based on the wrong branch (1.2.x, which is essentially vanilla apache 1.2.x). The correct target is the OpenHouse dev branch openhouse-1.2.0, which already has the id-bound ORC readers and the NestedField column-default API (#250) — so the id-binding and API backports here were redundant.

Rebuilt as a clean 2-PR stack on openhouse-1.2.0:

• ORC: fill initial defaults for missing id-bound fields #263 - ORC: fill initial defaults for missing id-bound fields
• Spark 3.1: fill ORC initial defaults on the row reader #264 - Spark 3.1: fill ORC initial defaults on the row reader