Address review: test naming, catalog-name no-op, silent-skip coverage

- Rename tests/table/test_delete_data_file_manifest_pruning_bug.py to
  test_snapshot_manifest_pruning.py, matching the repo's component-named
  test file convention (no _bug suffix).
- Drop the f"..._{catalog.name}" identifier suffix: all three catalog
  fixture params share name="test_catalog", so it was a no-op: isolation
  already comes from the per-test tmp_path.
- Add test_delete_data_file_manifest_pruning_bucket_on_same_result_type_succeeds:
  a BucketTransform over an IntegerType source column, where the pre-fix
  predicate's type happened to match the source column's type. Unlike the
  string-source case (a loud TypeError), this variant let the buggy
  row-domain predicate bind successfully and silently prune away the
  manifest containing the target file, so delete_data_file reported
  success without deleting anything. Verified this fails (silently, no
  exception) against the pre-fix code and passes with the current fix.

Author: QlikFrederic

…delete_data_file_manifest_pruning_bug.py …/table/test_snapshot_manifest_pruning.pytests/table/test_delete_data_file_manifest_pruning_bug.py renamed to tests/table/test_snapshot_manifest_pruning.py tests/table/test_delete_data_file_manifest_pruning_bug.py renamed to tests/table/test_snapshot_manifest_pruning.py

@@ -33,7 +33,7 @@ def test_delete_data_file_manifest_pruning_bucket_transform_succeeds(catalog: Ca
     works regardless of the partition transform.
     """
     catalog.create_namespace_if_not_exists("default")
-    identifier = f"default.bucket_delete_bug_{catalog.name}"
+    identifier = "default.bucket_delete"
 
     schema = Schema(
         NestedField(1, "tenant_id", StringType(), required=True),
@@ -95,7 +95,7 @@ def test_delete_data_file_manifest_pruning_predicate_uses_partition_field(catalo
     non-discriminating fallback. This test guards the pruning predicate itself.
     """
     catalog.create_namespace_if_not_exists("default")
-    identifier = f"default.bucket_delete_pruning_predicate_{catalog.name}"
+    identifier = "default.bucket_delete_pruning_predicate"
 
     schema = Schema(
         NestedField(1, "tenant_id", StringType(), required=True),
@@ -137,3 +137,51 @@ def test_delete_data_file_manifest_pruning_predicate_uses_partition_field(catalo
             predicate = overwrite._delete_files_partition_filters[existing_file.spec_id]
 
     assert predicate == EqualTo(Reference("tenant_id_bucket"), expected_bucket_id)
+
+
+def test_delete_data_file_manifest_pruning_bucket_on_same_result_type_succeeds(catalog: Catalog) -> None:
+    """delete_data_file must not silently skip a manifest when the bucket id happens to share the source column's type.
+
+    Pre-fix, the buggy predicate compared the source column (an int) against the bucket id
+    (also an int), so binding succeeded instead of raising. The manifest's min/max stats for
+    that column then incorrectly ruled out the manifest containing the target file, so the
+    whole manifest was skipped and the file was silently never deleted - no exception, no
+    error, just a delete that quietly did nothing. A string source column can't hit this path
+    since it would fail to bind (see the other tests here), so this needs a same-result-type
+    source to catch a regression back to the source-column domain.
+    """
+    catalog.create_namespace_if_not_exists("default")
+    identifier = "default.bucket_delete_same_result_type"
+
+    schema = Schema(NestedField(1, "value", IntegerType(), required=True))
+    spec = PartitionSpec(
+        PartitionField(
+            source_id=1,
+            field_id=1000,
+            transform=BucketTransform(8),
+            name="value_bucket",
+        ),
+        spec_id=0,
+    )
+    table = catalog.create_table(
+        identifier=identifier,
+        schema=schema,
+        partition_spec=spec,
+    )
+    table.append(
+        pa.Table.from_pylist(
+            [{"value": 42}],
+            schema=pa.schema([pa.field("value", pa.int32(), nullable=False)]),
+        )
+    )
+
+    before_paths = {task.file.file_path for task in table.scan().plan_files()}
+    existing_file = next(iter(table.scan().plan_files())).file
+
+    with table.transaction() as txn:
+        with txn.update_snapshot().overwrite() as overwrite:
+            overwrite.delete_data_file(existing_file)
+
+    after_paths = {task.file.file_path for task in table.scan().plan_files()}
+
+    assert before_paths - after_paths == {existing_file.file_path}