re-add after being lost in rebasing: make continuous writes reusable …

…if tests are run on an already existing etcd application

tests/integration/ha/helpers.py

@@ -28,12 +28,12 @@ async def existing_app(ops_test: OpsTest) -> str | None:
     return list(etcd_apps.keys())[0] if etcd_apps else None
 
 
-def start_continuous_writes(ops_test: OpsTest, endpoints: str, user: str, password: str) -> None:
+def start_continuous_writes(ops_test: OpsTest, app_name: str, endpoints: str, user: str, password: str) -> None:
     model = ops_test.model_full_name
     # this is the unit where the `etcdctl` command is executed
     # it does not mean that data is written to this cluster member
     # before removing the unit used for running `etcdctl`, continuous writes should be stopped
-    unit = ops_test.model.applications[APP_NAME].units[0].name
+    unit = ops_test.model.applications[app_name].units[0].name
     subprocess.Popen(
         [
             "python3",
@@ -52,9 +52,9 @@ def stop_continuous_writes() -> None:
     proc.communicate()
 
 
-def count_writes(ops_test: OpsTest, endpoints: str, user: str, password: str) -> str:
+def count_writes(ops_test: OpsTest, app_name: str, endpoints: str, user: str, password: str) -> str:
     model = ops_test.model_full_name
-    unit = ops_test.model.applications[APP_NAME].units[0].name
+    unit = ops_test.model.applications[app_name].units[0].name
     key = "cw_key"
 
     etcd_command = f"""{SNAP_NAME}.etcdctl \