Description
Is your feature request related to a problem? Please describe.
We're working with a client where we deal with large amount of data, even during development. For some parts we can't sample the data, and it makes the development process a bit painful. When running the pipeline, while in development mode, through elyra, sometimes we get some problems at the later steps of our own code. It's frustrating when you need to repeat the whole pipeline execution to test if a fix you did in the last step will work.
Describe the solution you'd like
It would be really nice to have a feature that allow us to just repeat the previous execution from the last failed step onwards. Kubeflow pipelines have it, and it usually works for us if we restart the job from kfp interface. However, when we change the code, we currently must restart from the start on elyra, becuase the .tar.gzs packages will need to generated again.
Describe alternatives you've considered
I understand that repeating from the last failed step is not something that applies to everyone. So, maybe we could have some sort of feature flag in the interface to enable/disable this kind of behavior. I think of it as a kind of development mode.
Additional context
Add any other context or screenshots about the feature request here.