fix:pyspark lazy load

pilot/connections/conn_spark.py

@@ -17,19 +17,19 @@ class SparkConnect(BaseConnect):
     driver: str = "spark"
     """db dialect"""
     dialect: str = "sparksql"
-    from pyspark.sql import SparkSession, DataFrame
 
     def __init__(
         self,
         file_path: str,
-        spark_session: Optional[SparkSession] = None,
+        spark_session: Optional = None,
         engine_args: Optional[dict] = None,
         **kwargs: Any,
     ) -> None:
         """Initialize the Spark DataFrame from Datasource path
         return: Spark DataFrame
         """
         from pyspark.sql import SparkSession
+
         self.spark_session = (
             spark_session or SparkSession.builder.appName("dbgpt_spark").getOrCreate()
         )
@@ -47,7 +47,7 @@ def from_file_path(
         except Exception as e:
             print("load spark datasource error" + str(e))
 
-    def create_df(self, path) -> DataFrame:
+    def create_df(self, path):
         """Create a Spark DataFrame from Datasource path(now support parquet, jdbc, orc, libsvm, csv, text, json.).
         return: Spark DataFrame
         reference:https://spark.apache.org/docs/latest/sql-data-sources-load-save-functions.html