Fix: Inconsistent serialization of query_cache in VectorDB

[PrinceSajjadHussain]

The VectorDB class's load_db method loads the query_cache as a JSON string and converts it to a dictionary using json.loads, but save_db uses json.dumps to store the query_cache dictionary back into a string format. This can be avoided for simplicity.

Applied fixes:
--- a/skills\text_to_sql\evaluation\vectordb.py
+++ b/skills\text_to_sql\evaluation\vectordb.py
@@ -13,7 +13,7 @@
if os.path.exists(self.db_path):
with open(self.db_path, "rb") as file:
data = pickle.load(file)

•        self.embeddings, self.metadata, self.query_cache = data['embeddings'], data['metadata'], json.loads(data['query_cache'])
  

•        self.embeddings, self.metadata, self.query_cache = data['embeddings'], data['metadata'], data['query_cache']
     else:
         self.embeddings, self.metadata, self.query_cache = [], [], {}
  

@@ -37,4 +37,4 @@

 def save_db(self):
     with open(self.db_path, "wb") as file:

•        pickle.dump({"embeddings": self.embeddings, "metadata": self.metadata, 
  

•                     "query_cache": json.dumps(self.query_cache)}, file)
  

•        pickle.dump({"embeddings": self.embeddings, "metadata": self.metadata, "query_cache": self.query_cache}, file)