fix(metadata): align GetCrossReference Thrift/SEA on parent identifiers

[eric-wang-1990]

Summary

GetCrossReference diverged between the Thrift and SEA protocols on the parent (PK-side) identifiers, and neither path matched the JDBC reference driver. Surfaced by the ADBC C# Thrift-vs-SEA comparator (get_cross_reference outcome diff): with an empty-string parent catalog + a valid foreign side (fk_child), Thrift threw while SEA returned rows.

Traced all four code paths for that input:

	Behavior (before)	JDBC reference
ADBC Thrift	throws	JDBC Thrift catches object-not-found → empty
ADBC SEA	returns rows (ignores parent)	JDBC SEA filters rows by parent → empty

Root causes & fixes

1. Thrift threw on an empty-string parent catalog. TGetCrossReferenceReq is null-guarded (HiveServer2Connection.cs) — a null parent is omitted (so the foreign-only GetColumnsExtended reuse works), but an empty-string parent is sent as ParentCatalogName="", which the server rejects (SHOW FOREIGN KEYS IN CATALOG \`→TABLE_OR_VIEW_NOT_FOUND/INVALID_PARAMETER_VALUE, verified live). JDBC Thrift (DatabricksThriftServiceClient.listCrossReferences) catches isObjectNotFoundExceptionand returns empty; ADBC Thrift had no such catch. → Catch object-not-found inDatabricksStatement.GetCrossReferenceAsync→ empty result.IsObjectNotFoundExceptiongains a static overload overAdbcExceptionso it matches theHiveServer2Exceptionthe Thrift path throws (same pattern asIsDescTableExtendedUnsupported`).

2. SEA ignored the parent identifiers entirely. SHOW FOREIGN KEYS is scoped to the foreign table, returning FKs to every parent; the JDBC reference filters those rows by the requested parent (CrossReferenceKeysDatabricksResultSetAdapter.includeRow), but ADBC SEA did not — a latent over-return bug for any table with FKs to multiple parents.
→ Filter returned rows by any specified (non-null) parent catalog/schema/table in GetCrossReferenceAsyncNoThrow. A null parent means "no constraint" (preserves the GetColumnsExtended foreign-only reuse, which passes null on all three); an empty-string parent matches only an empty row value → filters to empty.

Both paths now return an empty result for an empty-string parent + valid foreign side (matching JDBC spec semantics — an empty-string identifier names no real object), and SEA no longer over-returns for multi-parent tables.

Test Plan

• IsObjectNotFoundException static overload over HiveServer2Exception — empty-parent error shapes (TABLE_OR_VIEW_NOT_FOUND/42P01, INVALID_PARAMETER_VALUE) return true; unrelated (ACCESS_DENIED) returns false.
• ParentMatches — null = no filter (matches any, incl. empty row); exact + case-insensitive match; empty-string and non-matching parent filter out.
• Full unit suite: 974 passed, 0 failed, 0 skipped.
• Comparator re-run confirms the get_cross_reference diff collapses (supersedes driver-test PR #1249, which whitelisted it).

Supersedes the whitelist approach in databricks-driver-test #1249 with a real driver fix.

This pull request and its description were written by Isaac.

[peco-review-bot]

Verdict: 1 Low

Looks good — a focused, well-tested Thrift/SEA parity fix for GetCrossReference. The static IsObjectNotFoundException(AdbcException) overload, the Thrift object-not-found catch, and the SEA ParentMatches filter all line up with the JDBC reference semantics, and the unit tests cover the key cases. One low-severity edge case noted inline around the null-parent-column fallback in the SEA filter.

[peco-review-bot]

Verdict: 2 Low

Looks good — a focused, correct alignment of GetCrossReference across Thrift/SEA that matches the JDBC reference (verified CrossReferenceKeysDatabricksResultSetAdapter.includeRow), with solid unit coverage of the two new helpers. Two Low notes: the Thrift catch broadens object-not-found swallowing to the FK side, and the new filter/catch wiring is only covered indirectly (predicate + classifier), not end-to-end.

[peco-review-bot]

Verdict: 1 Low

Looks good — a clean, JDBC-parity-aligned fix with strong unit coverage (predicate + end-to-end Http-seam tests). Verified the two GetCrossReferenceAsyncNoThrow call sites pass parents correctly (null for the GetColumnsExtended reuse, distinct parent/foreign options for the metadata path), that ShouldReturnEmptyPKFKResult does not short-circuit the empty-string-parent case, and that the static IsObjectNotFoundException overload + narrowly-guarded Thrift catch behave as described. One low note on null/absent server parent columns potentially over-filtering on SEA only.