refactor: drop empty-batch workaround in MetadataResultSets

Now that ArrowStreamReaderCursor.loadNextNonEmptyBatch (introduced
earlier on this branch as a pre-unify cursor fix) consumes empty
batches at the cursor seam, MetadataResultSets.writeArrowStream no
longer needs its own "skip writeBatch when rowCount==0" workaround:
the cursor handles the empty-only case correctly. Remove the special
case and always emit a batch.

Tightens the zeroRowOnlyBatchYieldsNoRows test docstring to match.

Author: mkaufmann

jdbc-core/src/main/java/com/salesforce/datacloud/jdbc/core/metadata/MetadataResultSets.java

@@ -96,12 +96,7 @@ private static byte[] writeArrowStream(List<ColumnMetadata> columns, List<List<O
             ByteArrayOutputStream out = new ByteArrayOutputStream();
             try (ArrowStreamWriter writer = new ArrowStreamWriter(root, null, out)) {
                 writer.start();
-                // Skip writeBatch() for empty results — writing a zero-row batch confuses the
-                // cursor (see ArrowStreamReaderCursor.next), which interprets a successfully
-                // loaded batch as "at least one row available".
-                if (root.getRowCount() > 0) {
-                    writer.writeBatch();
-                }
+                writer.writeBatch();
                 writer.end();
             }
             return out.toByteArray();

jdbc-core/src/test/java/com/salesforce/datacloud/jdbc/core/ArrowStreamReaderCursorTest.java

@@ -64,6 +64,10 @@ void incrementsInternalIndexUntilRowsExhaustedThenLoadsNextBatch() {
         val sut = new ArrowStreamReaderCursor(reader, allocator, ZoneId.systemDefault());
         IntStream.range(0, times + 1).forEach(i -> sut.next());
 
+        // Each next() inspects rowCount once on the per-batch index check. loadNextNonEmptyBatch
+        // is reached on the (times+1)-th call but only inspects rowCount inside its loop body if
+        // loadNextBatch returns true; here it returns false, so getRowCount is observed times+1
+        // times in total.
         verify(root, times(times + 1)).getRowCount();
         verify(reader, times(1)).loadNextBatch();
     }