Add `RunnableMixin` for pre/post processing on arbitrary run execution

[vachillo]

• [x] I have read and agree to the contributing guidelines.

Is your feature request related to a problem? Please describe. there is a lot of different areas of the framework that could benefit from pre/post processing hooks on data.

Describe the solution you'd like implement a RunnableMixin:

class RunnableMixin:
process_before_run_fn: Callable[[Any], Any]
process_after_run_fn: Callable[[Any], Any]

def before_run(self, *args, **kwargs) -> None (or Any?): ....
def run(self, *args, **kwargs) -> Any: ...
def after_run(self, *args, **kwargs) -> None (or Any?): ...

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context Add any other context or screenshots about the feature request here.

[collindutter]

I'd like to hear more about what kind of pre/post processing you're envisioning. Is it mostly for data coercion of LLM output?

[vachillo]

I'd like to hear more about what kind of pre/post processing you're envisioning. Is it mostly for data coercion of LLM output?

Mostly just getting the output into a format that is either a more specific type of artifact, or a different Python type.

As an example, I have a basic PromptTask that I am asking the LLM "respond with only true or false". I want to parse the output into a BooleanArtifact right on the task, so on the next task I can be confident what the type is.

[vachillo]

I can also see this being a way to mask/unmask data at a granular level.