[TypeTransformer] Support non-Any Python types as Any input in workflows

[Future-Outlier]

Tracking issue

flyteorg/flyte#5366

Why are the changes needed?

We want to allow the Any type to accept all kinds of inputs and outputs, not just the types we currently can't handle.

For example, workflow like this.

@task
def foo(a: Any) -> int:
    if type(a) == int:
        return a + 1
    return 0

@workflow
def wf(a: int) -> int:
    return foo(a=a)

Note: This PR only supports Any in python, if we want to support java, we will need to add more code.

What changes were proposed in this pull request?

Check if the literal's scalar has a primitive type.
If it does, convert it to a Python value using the corresponding type transformer.

How was this patch tested?

unit test and remote cluster

from flytekit import task, workflow
from typing import Any

@task
def foo(a: Any) -> int:
    if type(a) == int:
        return a + 1
    return 0

@workflow
def wf(a: int) -> int:
    return foo(a=a)

FROM python:3.9-slim-buster
USER root
WORKDIR /root
ENV PYTHONPATH /root
RUN apt-get update && apt-get install build-essential -y
RUN apt-get install git -y

RUN pip install -U git+https://github.com/flyteorg/flytekit.git@3804a5155a523e2028ece8e7e581f794ebabd788

Setup process

pyflyte run --remote --image localhost:30000/any:0522 any_task.py wf --a 1

Screenshots

Check all the applicable boxes

• I updated the documentation accordingly.
• All new and existing tests passed.
• All commits are signed-off.

Related PRs

flyteorg/flyte#5408

[Future-Outlier]

Update: I'm not sure if it's ok to support primitive types only.

[Future-Outlier]

It can only run in local execution and pyflyte run --remote, since I haven't discovered a way to create metadata at flyteconsole.

@dataclass
class Datum:
    x: int
    y: str
    z: dict[int, str]
    w: list[int]
    
  @task
def foo_dataclass(a: Datum) -> int:
    if type(a) == Datum:
        return a.x
    return 0

@workflow
def foo_wf(a: Datum) -> int:
    return foo_dataclass(a=a)

pyflyte run --remote --image localhost:30000/any:0528-1445 any_task.py foo_wf --a '{ "x": 1, "y": "2", "z": { "1": "one", "2": "two" }, "w": [1, 2, 3] }'

[Future-Outlier]

Note:
python dotted path can solve type for Dataclass, while eval can solve most of python type,
we should use both of these ways