[ray-data] Add alias parameters to the aggregate function, and add quantile fn

[yiwei00000]

Hi:
I added the alias parameter in the aggregate function and returned the alias. At the same time, I added aggregate function, quantile .

The following is an example of usage :

from ray.data.aggregate import Max, Mean, Min, Quantile, Std
import ray

ls = [
{"A": i, "B": i ** 2}
for i in range(10)]
ds = ray.data.from_items(ls)

ops = [Max("A", alias_name="max_a"), Min("A", alias_name="min_a"), Mean("A", alias_name="mean_a"),
Std("A", alias_name="std_a"),Quantile("A", percent=0.5, alias_name="quantile_a")]
print(dict(ds.aggregate(*ops)))

rs:
{'max_a': 9, 'min_a': 0, 'mean_a': 4.5, 'std_a': 3.0276503540974917, 'quantile_a': 4.5}

[ericl]

API +1 from me, cc @c21 @scottjlee

[scottjlee]

LGTM, just a couple small nits. Thanks!

[yiwei00000]

Hi，is there any other issues? I will cooperate and correct it in a timely manner. @scottjlee @ericl @scv119

[ericl]

Lgtm. I think there's one unresolved comment from Scott above still.

[ericl]

Looks like there's also a merge conflict with master

[yiwei00000]

Looks like there's also a merge conflict with master

Yes, I don't know if there is a file name conflict. I merged the upstream/master and the conflict was resolved. @ericl

[ericl]

Merged, thanks!

[yiwei00000]

Merged, thanks!

Thank you very much. I still have some areas to add or modify in the future, including APIs or new features.