feat(submit): require explicit confirmation for all prepare_submit_job params

Remove defaults from queue, executor_memory, executor_cores,
num_executors, and app_name in prepare_submit_job. All must now be
explicitly confirmed by the caller.

Also add extra_args to the PendingSubmission snapshot so users can
confirm non-conf spark-submit flags (e.g. --jars, --py-files) at
prepare time; confirm_submit_job passes them through to
build_spark_submit_command.

This is an intentional breaking change to the MCP tool contract:
callers can no longer rely on implicit defaults.

tests/unit/test_pending_store.py

@@ -11,6 +11,7 @@ from spark_executor.models import PendingSubmission
 def _pending(pid: str = "p_1", created_at: datetime | None = None) -> PendingSubmission:
     return PendingSubmission(
         pending_id=pid,
+        app_name="test",
         connection="prod",
         master="yarn",
         deploy_mode="cluster",
@@ -20,6 +21,7 @@ def _pending(pid: str = "p_1", created_at: datetime | None = None) -> PendingSub
         executor_cores=2,
         num_executors=2,
         spark_conf={},
+        extra_args={},
         created_at=created_at or datetime(2026, 6, 24),
     )
 
@@ -80,10 +82,10 @@ def test_delete_removes_from_date_file_and_deletes_empty_file():
 
 def _legacy_record_text() -> str:
     return (
-        '{"p_legacy": {"pending_id": "p_legacy", "connection": "prod", '
-        '"master": "yarn", "deploy_mode": "cluster", "script_path": "/tmp/j.py", '
-        '"queue": "default", "executor_memory": "4G", "executor_cores": 2, '
-        '"num_executors": 2, "spark_conf": {}, '
+        '{"p_legacy": {"pending_id": "p_legacy", "app_name": "legacy", '
+        '"connection": "prod", "master": "yarn", "deploy_mode": "cluster", '
+        '"script_path": "/tmp/j.py", "queue": "default", "executor_memory": "4G", '
+        '"executor_cores": 2, "num_executors": 2, "spark_conf": {}, '
         '"created_at": "2026-06-23T00:00:00"}}'
     )