Add source & stage decorators

[dagardner-nv]

Description

• Adds two new decorators source and stage allowing stages to be created from python functions.
• Accept & return types are inferred by inspecting the type annotations of the decorated function.
• This removes the need for several lines of boilerplate code needed for simple Python-only stages.
• Includes unrelated docstring fix to morpheus/pipeline/stage_base.py

example:

from morpheus.pipeline.stage_decorator import source
from morpheus.pipeline.stage_decorator import stage

@source
def source_gen(dataframes: list[cudf.DataFrame]) -> collections.abc.Iterator[MessageMeta]:
    for df in dataframes:
        yield MessageMeta(df)


@stage
def on_data(message: MessageMeta) -> MessageMeta:
    with message.mutable_dataframe() as df:
        print(df)
    return message

Closes #1253

By Submitting this PR I confirm:

• I am familiar with the Contributing Guidelines.
• When the PR is ready for review, new or existing tests cover these changes.
• When the PR is ready for review, the documentation is up to date with these changes.

[mdemoret-nv]

This is a good start but we need some changes to be forward looking with these decorators. General changes:

1. The classes used by the decorators should be "dumber" and move most of the implementation into the decorator instead of the class. See my comments for more details.
2. Need some tests validating the parameters used at construction time
3. Needs documentation showing how this can be used in the developer guide.

[mdemoret-nv]

Looks good now. Like the changes. This will be super useful in the future.

[mdemoret-nv]

/merge

[mdemoret-nv]

/merge