`@pipe_output`, `@pipe_input` and `@mutate` for async functons

[elijahbenizzy]

Is your feature request related to a problem? Please describe. These don't work:

1. Async functions decorated with @pipe_output/@pipe_input/@mutate (have not tested all configurations of these)
2. Async transformations covered by step

Describe the solution you'd like These should work. See the way we do other decorators. E.G. https://github.com/DAGWorks-Inc/hamilton/blob/9ae4183e8ceb98f3393c8d62923fb205cc370eb0/hamilton/function_modifiers/recursive.py#L399

Describe alternatives you've considered We can use a workaround in some cases. E.G. an identity node. See the first comment for it.

Additional context https://hamilton-opensource.slack.com/archives/C03M33QB4M8/p1729156532663429

[elijahbenizzy]

Workaround for @pipe_output

import asyncio

import pandas as pd

from hamilton import async_driver
from hamilton.function_modifiers import pipe_output, step, hamilton_exclude

async def data_input() -> pd.DataFrame:
    await asyncio.sleep(0.0001)
    return pd.DataFrame({
        "a": [1, 2, 3],
        "b": [4, 5, 6]
    })

def _groupby_a(d: pd.DataFrame) -> pd.DataFrame:
    return d.groupby("a").sum().reset_index()

def _groupby_b(d: pd.DataFrame) -> pd.DataFrame:
    return d.groupby("b").sum().reset_index()

@pipe_output(
    step(_groupby_a).when(groupby="a"),
    step(_groupby_b).when_not(groupby="a"),
)
def data(data_input: pd.DataFrame) -> pd.DataFrame:
    return data_input

@hamilton_exclude
async def main():
    import __main__
    dr = (await async_driver.Builder().with_modules(__main__).with_config(dict(groupby="b")).build())
    results = await dr.execute(["data"])
    print(results)

if __name__ == "__main__":
    asyncio.run(main())