BUG/PERF: DataFrame.isin lossy data conversion

[lukemanley]

• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/v2.1.0.rst file if fixing a bug or adding a new feature.

fixes edge case bug (lossy data conversion):

import pandas as pd

val = 1666880195890293744

df = pd.DataFrame({"a": [val], "b": [1.0]})

df.apply(lambda x: x.isin([val]))  # True, False
df.isin([val])                     # False, False 

perf (for EA dtypes):

import pandas as pd
import numpy as np

data = np.random.randint(0, 1000, (100_000, 2))
df = pd.DataFrame(data, dtype="int64[pyarrow]")

%timeit df.isin([1, 50])

# 25.6 ms ± 1.49 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)  <- main
# 2.1 ms ± 265 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)   <- PR

[mroeschke]

Thanks @lukemanley