MappedTasks overwrite each other #6
Description
When you run a FlowpyterOperator as a MappedTask, the last task to complete is the notebook that lands in executed_notebooks, overwriting any other executed notebooks of that task. If the number of MappedTasks expands after a task has been run a few times, this also means that earlier notebooks get rewritten with the new tasks as they run with run_id=1 ect.