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.

spark_executor/models.py

@@ -86,7 +86,7 @@ class Connection(BaseModel):
 
 class PendingSubmission(BaseModel):
     pending_id: str
-    app_name: str | None = None
+    app_name: str
     connection: str
     master: str
     deploy_mode: str
@@ -97,6 +97,7 @@ class PendingSubmission(BaseModel):
     executor_cores: int
     num_executors: int
     spark_conf: dict[str, str] = Field(default_factory=dict)
+    extra_args: dict[str, str] = Field(default_factory=dict)
     created_at: datetime
     status: str = "PENDING"  # PENDING | SUBMITTED | CANCELLED | FAILED
     error: str | None = None

spark_executor/tools/requests.py

@@ -33,9 +33,9 @@ class SaveConnectionRequest(BaseModel):
 
 class PrepareSubmitJobRequest(BaseModel):
     connection: str
-    app_name: str | None = Field(
-        default=None,
-        description="Optional human-readable application name for tracking the pending submission.",
+    app_name: str = Field(
+        ...,
+        description="Human-readable application name for tracking the pending submission.",
     )
     script_path: str = Field(
         ...,
@@ -49,10 +49,26 @@ class PrepareSubmitJobRequest(BaseModel):
             "if the path is missing or not a file."
         ),
     )
-    queue: str = "default"
-    executor_memory: str = "4G"
-    executor_cores: int = 2
-    num_executors: int = 2
+    queue: str = Field(
+        ...,
+        description="YARN queue to submit to. Must be explicitly confirmed by the caller.",
+    )
+    executor_memory: str = Field(
+        ...,
+        description="Executor memory, e.g. '4G'. Must be explicitly confirmed by the caller.",
+    )
+    executor_cores: int = Field(
+        ...,
+        description="Number of cores per executor. Must be explicitly confirmed by the caller.",
+    )
+    num_executors: int = Field(
+        ...,
+        description="Total number of executors. Must be explicitly confirmed by the caller.",
+    )
+    extra_args: dict[str, str] | None = Field(
+        default=None,
+        description="Additional spark-submit flags (e.g. jars, py-files) confirmed at prepare time.",
+    )
 
 
 class PendingIdRequest(BaseModel):

spark_executor/tools/submit.py

@@ -64,17 +64,18 @@ def prepare_submit_job(
     *,
     connection: str,
     script_path: str,
-    queue: str = "default",
-    executor_memory: str = "4G",
-    executor_cores: int = 2,
-    num_executors: int = 2,
-    app_name: str | None = None,
+    queue: str,
+    executor_memory: str,
+    executor_cores: int,
+    num_executors: int,
+    app_name: str,
+    extra_args: dict[str, str] | None = None,
 ) -> dict[str, object]:
     """Snapshot connection params and persist a PendingSubmission. Does NOT submit."""
     logger.debug(
         f"prepare_submit_job enter connection={connection} script_path={script_path} "
         f"queue={queue} executor_memory={executor_memory} executor_cores={executor_cores} "
-        f"num_executors={num_executors}"
+        f"num_executors={num_executors} app_name={app_name}"
     )
     # Order of checks matters for the error the agent sees:
     #   1. Unknown connection  -> 404 (KeyError -> 404 in server.py)
@@ -121,6 +122,7 @@ def prepare_submit_job(
         executor_cores=executor_cores,
         num_executors=num_executors,
         spark_conf=dict(conn.spark_conf),
+        extra_args=dict(extra_args or {}),
         created_at=datetime.utcnow(),
         status="PENDING",
     )
@@ -164,6 +166,7 @@ def confirm_submit_job(*, pending_id: str) -> SubmitResult:
         executor_cores=pending.executor_cores,
         num_executors=pending.num_executors,
         spark_conf=pending.spark_conf,
+        extra_args=pending.extra_args,
     )
     logger.info(
         f"confirm_submit_job start pending_id={pending_id} "