Refactor scripts for vector index and search matrix summarization

- Updated dataset and resource configurations in `run_11_vector_index_build_matrix.sh` and `run_12_vector_search_matrix.sh` to use larger datasets and increased resource limits.
- Introduced normalization functions for run labels in multiple scripts to ensure consistent labeling across runs.
- Enhanced summary scripts (`summarize_07_tables_oltp_matrix.sh`, `summarize_08_tables_olap_matrix.sh`, `summarize_09_graph_oltp_matrix.sh`, `summarize_10_graph_olap_matrix.sh`, `summarize_11_vector_index_build_matrix.sh`, `summarize_12_vector_search_matrix.sh`) to include additional metrics and improve data handling.
- Implemented new functions for handling hash groups and row counts in OLAP summary scripts to facilitate better reporting of cross-DB comparisons.
- Improved error handling and directory management in disk usage calculations.

Author: tae898

bindings/python/examples/02_social_network_graph.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-ArcadeDB Python Bindings - Social Network Graph Example
+Example 02: Social Network Graph
 
 This example demonstrates how to use ArcadeDB as a graph database to model
 a social network with people and friendships. It showcases:
@@ -1166,6 +1166,6 @@ def print_section_header(title, emoji="🔹"):
 
 
 if __name__ == "__main__":
-    print("🌐 ArcadeDB Python - Social Network Graph Example")
+    print("🌐 ArcadeDB Python - Example 02: Social Network Graph")
     print("=" * 55)
     main()

bindings/python/examples/03_vector_search.py

@@ -48,7 +48,9 @@
 import numpy as np
 
 # Parse command line arguments
-parser = argparse.ArgumentParser(description="Vector Search Example")
+parser = argparse.ArgumentParser(
+    description="Example 03: Vector Search - Semantic Similarity"
+)
 args = parser.parse_args()
 
 print("=" * 70)

bindings/python/examples/04_csv_import_documents.py

@@ -896,7 +896,7 @@ def check_dataset_exists(data_dir):
 
 # Parse command-line arguments
 parser = argparse.ArgumentParser(
-    description="Import MovieLens dataset into ArcadeDB",
+    description="Example 04: Import MovieLens dataset into ArcadeDB",
     formatter_class=argparse.RawDescriptionHelpFormatter,
     epilog="""
 Examples:

bindings/python/examples/05_csv_import_graph.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-Graph Creation Benchmark - Clean Architecture
+Example 05: Graph Creation Benchmark - Clean Architecture
 
 This benchmark compares graph creation strategies with multiple options:
 - Method: Java API vs SQL
@@ -2152,7 +2152,7 @@ def run_and_validate_queries(db: Any, size: str, check_baseline: bool = True):
 
 
 def main():
-    parser = argparse.ArgumentParser(description="Graph Creation Benchmark")
+    parser = argparse.ArgumentParser(description="Example 05: Graph Creation Benchmark")
     parser.add_argument(
         "--dataset",
         choices=["movielens-small", "movielens-large"],
@@ -2231,7 +2231,7 @@ def main():
         db_name = args.db_name
 
     print("=" * 70)
-    print("🚀 Graph Creation Benchmark")
+    print("🚀 Example 05: Graph Creation Benchmark")
     print("=" * 70)
     print(f"Dataset: {args.dataset}")
     print(f"Batch size: {args.batch_size:,}")

bindings/python/examples/06_vector_search_recommendations.py

@@ -425,7 +425,9 @@ def vector_based_recommendations(
 
 
 def main():
-    parser = argparse.ArgumentParser(description="Vector Search Movie Recommendations")
+    parser = argparse.ArgumentParser(
+        description="Example 06: Vector Search Movie Recommendations"
+    )
 
     parser.add_argument(
         "--db-path",

bindings/python/examples/07_stackoverflow_tables_oltp.py

@@ -577,10 +577,8 @@ def create_schema_duckdb(conn):
 
 
 def create_duckdb_id_indexes(conn):
-    for table in TABLE_DEFS:
-        conn.execute(
-            f'CREATE INDEX IF NOT EXISTS idx_{table["name"].lower()}_id ON "{table["name"]}"("Id")'
-        )
+    print("Skipping manual DuckDB secondary indexes for this benchmark.")
+    return 0.0
 
 
 def create_schema_postgresql(conn):
@@ -1433,9 +1431,7 @@ def run_oltp_duckdb(
         f"Ingest end   (duckdb, UTC): {ingest_ended_at} "
         f"(elapsed={preload_time:.2f}s)"
     )
-    index_start = time.time()
-    create_duckdb_id_indexes(conn)
-    index_time = time.time() - index_start
+    index_time = create_duckdb_id_indexes(conn)
 
     load_counts_start = time.time()
     preload_counts = count_table_rows_sql(conn)
@@ -2155,7 +2151,9 @@ def run_in_docker(args):
 
 
 def main():
-    parser = argparse.ArgumentParser(description="Stack Overflow Tables (OLTP)")
+    parser = argparse.ArgumentParser(
+        description="Example 07: Stack Overflow Tables (OLTP)"
+    )
     parser.add_argument(
         "--dataset",
         choices=sorted(EXPECTED_DATASETS),

bindings/python/examples/08_stackoverflow_tables_olap.py

@@ -998,17 +998,8 @@ def create_indexes_sqlite(conn: sqlite3.Connection) -> float:
 
 
 def create_indexes_duckdb(conn) -> float:
-    start = time.time()
-    for table, props, unique in INDEX_DEFS:
-        suffix = "_".join(props)
-        index_name = f"idx_{table}_{suffix}"
-        unique_sql = "UNIQUE " if unique else ""
-        ddl = (
-            f"CREATE {unique_sql}INDEX IF NOT EXISTS {index_name} "
-            f"ON {table} ({', '.join(props)})"
-        )
-        conn.execute(ddl)
-    return time.time() - start
+    print("Skipping manual DuckDB secondary indexes for this benchmark.", flush=True)
+    return 0.0
 
 
 def create_indexes_postgresql(conn) -> float:
@@ -1561,10 +1552,7 @@ def run_olap_arcadedb(
         },
         "index": {
             "total_s": index_elapsed,
-            "indexes": [
-                {"table": t, "properties": props, "unique": unique}
-                for t, props, unique in INDEX_DEFS
-            ],
+            "indexes": [],
         },
         "queries": {
             "total_s": query_elapsed,
@@ -2002,7 +1990,7 @@ def query_runner(sql: str) -> List[Dict[str, Any]]:
 
 def main():
     parser = argparse.ArgumentParser(
-        description="Stack Overflow Tables (OLAP)",
+        description="Example 08: Stack Overflow Tables (OLAP)",
     )
     parser.add_argument(
         "--dataset",