[default values] ORC: don't push file filters for omitted initial-default fields

[cbb330]

Summary

• Exact copy of #269 onto openhouse-1.5.2.
• Same omit-from-SARG rule, same TypeUtil NestedField.from default preservation, same Spark scan/reader tests.

Required 1.5.2-only delta: ExpressionToSearchArgument.predicate already skipped non-BoundReference terms on this line; kept that conjunct and added the omitted-field-id check. API/ORC tests in this tree are JUnit 5/AssertJ (JUnit 4 Assert is not on those classpaths). Spark tests stay JUnit 4, copied into spark/v3.5.

Stacked on #275 (the #274 retarget / exact copy of #268). Both PRs target openhouse-1.5.2. Merge 275 first; this PR then shows only the #269 commit.

Testing Done

export JAVA_HOME=$(/usr/libexec/java_home -v 17)
./gradlew -DsparkVersions=3.5 -DscalaVersion=2.12 -DhiveVersions= -DflinkVersions= \
  :iceberg-api:test --tests 'org.apache.iceberg.types.TestTypeUtil.testAssignFreshIdsPreservesDefaults' \
  :iceberg-orc:test --tests 'org.apache.iceberg.orc.TestExpressionToSearchArgument.testDisables*' \
  --tests 'org.apache.iceberg.orc.TestExpressionToSearchArgument.testMixedPhysicalAndOmittedDefaultPredicates' \
  :iceberg-spark:iceberg-spark-3.5_2.12:test \
  --tests 'org.apache.iceberg.spark.source.TestSparkOrcInitialDefaultScan' \
  --tests 'org.apache.iceberg.spark.data.TestSparkOrcReaderForFieldsWithDefaultValue' \
  -x generateGitProperties