error on _resolve_should_track_driver_status with spark standalone (cluster mode)

### Apache Airflow Provider(s) apache-spark ### Versions of Apache Airflow Providers apache-airflow-providers-apache-spark 5.0.0 pyspark 3.5.4 ### Apache Airflow version 2.10.3 ### Operating System ubuntu server 22.04 ### Deployment Other ### Deployment details NA ### What happened Hi SparkSubmitOperator fails when interacting with a spark standalone server in deploy-mode=cluster. In such condition, the _resolve_should_track_driver_status function in /usr/local/lib/python3.12/site-packages/airflow/providers/apache/spark/hooks/spark_submit.py returns True... I have exactly the same pb as the one discussed here : https://github.com/apache/airflow/discussions/21799 I did find a case opened for this issue. ### What you think should happen instead I have an error when _resolve_should_track_driver_status returns True. If I modify the function to return False, testing the task passes but in such case the final spark job status is not confirmed by the spark driver. ### How to reproduce ``` cat sparkoperator_test.py import airflow from datetime import datetime, timedelta from airflow.providers.apache.spark.operators.spark_submit import SparkSubmitOperator default_args = { 'owner': 'xxxxxxxx', 'start_date': datetime(2025, 1, 31), } with airflow.DAG('sparkoperator_test', default_args=default_args, schedule_interval='@hourly', description='test of sparkoperator in airflow by Omar and Regis', ) as dag: spark_conf={ 'spark.standalone.submit.waitAppCompletion':'true', } spark_compute_pi = SparkSubmitOperator( task_id='spark-compute-pi', conn_id='spark_qa1', java_class='org.apache.spark.examples.SparkPi', application='/opt/spark/examples/jars/spark-examples_2.12-3.5.3.jar', name='compute_pi_from_airflow', application_args=["30"], total_executor_cores=4, executor_m…