[BUG]Failed to run execute(wait=False) in another thread

[VessaliusOz]

Describe the bug
execute(wait=False) failed when running in a thread pool

import os
import time
import mars.tensor as mt
import mars.dataframe as md
import concurrent.futures


def func():
    df = md.DataFrame(
            mt.random.rand(4, 4, chunk_size=2),
            columns=list("abcd"),
        )
    df.execute()
    pid = os.getpid()
    info = df.apply(
        lambda s: s if os.getpid() == pid else (time.sleep(5) or s)
    ).execute(wait=False)
    result = info.result()
    return result


with concurrent.futures.ThreadPoolExecutor(max_workers=1) as executor:
    future = executor.submit(func)
    print(future.result())

To Reproduce
To help us reproducing this bug, please provide information below:

1. Your Python version 3.9.13
2. The version of Mars you use 0.9.0
3. Versions of crucial packages, such as numpy, scipy and pandas
   i. numpy: 1.23.0
   ii. ray: 1.13.0
   iii. scipy: 1.8.1
   iv. pandas: 1.4.3
4. Full stack of the error.

Traceback (most recent call last):
  File "/Users/vessalius/.local/share/virtualenvs/outer_mars-bUMhzz2w/lib/python3.9/site-packages/mars/deploy/oscar/session.py", line 101, in _ensure_future
    self._future_local.future
AttributeError: '_thread._local' object has no attribute 'future'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/vessalius/Desktop/ray_serving_test/problem.py", line 25, in <module>
    print(future.result())
  File "/usr/local/Cellar/python@3.9/3.9.13_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/concurrent/futures/_base.py", line 446, in result
    return self.__get_result()
  File "/usr/local/Cellar/python@3.9/3.9.13_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/concurrent/futures/_base.py", line 391, in __get_result
    raise self._exception
  File "/usr/local/Cellar/python@3.9/3.9.13_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/Users/vessalius/Desktop/ray_serving_test/problem.py", line 19, in func
    result = info.result()
  File "/Users/vessalius/.local/share/virtualenvs/outer_mars-bUMhzz2w/lib/python3.9/site-packages/mars/deploy/oscar/session.py", line 131, in result
    self._ensure_future()
  File "/Users/vessalius/.local/share/virtualenvs/outer_mars-bUMhzz2w/lib/python3.9/site-packages/mars/deploy/oscar/session.py", line 110, in _ensure_future
    self._future_local.aio_future = asyncio.wrap_future(fut)
  File "/usr/local/Cellar/python@3.9/3.9.13_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/asyncio/futures.py", line 411, in wrap_future
    loop = events.get_event_loop()
  File "/usr/local/Cellar/python@3.9/3.9.13_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/asyncio/events.py", line 642, in get_event_loop
    raise RuntimeError('There is no current event loop in thread %r.'
RuntimeError: There is no current event loop in thread 'ThreadPoolExecutor-1_0'.

5. Minimized code to reproduce the error.

Expected behavior
A clear and concise description of what you expected to happen.

Additional context
Add any other context about the problem here.

[fyrestone]

The ExecutionInfo.result() does not have to call asyncio.wrap_future. I will fix it.