Add append delta filename to metadata.append_deltas view

Why these changes are being introduced:
* The method for merging append deltas into the static metadata database file
needs the filenames of append deltas to easily identify which files
to delete (once merged). Prior to this change, the append deltas view
only had a 'filename' column, which referred to the path or S3 URI for
the TIMDEXDataset parquet file.

How this addresses that need:
* Set filename='append_delta_filename' when creating metadata.append_deltas view
* Explicitly select metadata column names when creating metadata.records view

Side effects of this change:
* None

Relevant ticket(s):
* https://mitlibraries.atlassian.net/browse/TIMX-528

Author: jonavellecuerdo

timdex_dataset_api/metadata.py

@@ -396,7 +396,8 @@ def _create_append_deltas_view(self, conn: DuckDBPyConnection) -> None:
             create or replace view metadata.append_deltas as (
                 select *
                 from read_parquet(
-                    '{self.append_deltas_path}/*.parquet'
+                    '{self.append_deltas_path}/*.parquet',
+                    filename = 'append_delta_filename'
                 )
             );
             """
@@ -414,14 +415,17 @@ def _create_append_deltas_view(self, conn: DuckDBPyConnection) -> None:
 
     def _create_records_union_view(self, conn: DuckDBPyConnection) -> None:
         logger.debug("creating view of unioned records")
+
         conn.execute(
-            """
+            f"""
             create or replace view metadata.records as
             (
-                select *
+                select
+                    {','.join(ORDERED_METADATA_COLUMN_NAMES)}
                 from static_db.records
                 union all
-                select *
+                select
+                    {','.join(ORDERED_METADATA_COLUMN_NAMES)}
                 from metadata.append_deltas
             );
             """