Union types don't recognize bools in coercion order if ints or floats are included

[dlopuch]

Checks

• I added a descriptive title to this issue
• I have searched (google, github) for similar issues and couldn't find anything
• I have read and followed the docs and still think this is a bug

Bug

Output of python -c "import pydantic.utils; print(pydantic.utils.version_info())":

             pydantic version: 1.8.1
            pydantic compiled: True
                 install path: ~/.local/share/virtualenvs/XXX-D1SziJ_u/lib/python3.6/site-packages/pydantic
               python version: 3.6.13 (default, Feb 18 2021, 16:38:18)  [GCC Apple LLVM 12.0.0 (clang-1200.0.32.29)]
                     platform: Darwin-20.3.0-x86_64-i386-64bit
     optional deps. installed: ['typing-extensions']

This is kinda related to #2135, but a different enough case that it probably warrants its own issue.

In that bug, it was said "pydantic tries to coerce in the order of the union". The docs say something similar: "pydantic will attempt to 'match' any of the types defined under Union and will use the first one that matches."

That behavior appears to not work for bools -- bools seem to be cooerced to ints or floats if int or float is defined in the list:

import typing as t
from pydantic import BaseModel

class UnionTest(BaseModel):
     mixed: t.Optional[ t.Union[bool, int, str] ]

UnionTest(mixed=False)
# ===> UnionTest(mixed=0) -- BUG: coerced into int, not the bool!

[PrettyWood]

@dlopuch If you can use python 3.7+ you won't have this issue
This works in 3.6 and fails with 3.7+

from typing import Union

assert Union[bool, int, str] == Union[int, str]

PS: As a workaround for python 3.6, you can use StrictBool