BUG: Replace raises TypeError if to_replace is Dict with numeric DataFrame and key of Dict is String

[phofl]

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

import pandas as pd

df = pd.DataFrame({"Y0": [1, 2], "Y1": [3, 4]})
df = df.replace({"replace_string": "test"})

Problem description

This raises a TypeError with the following Traceback:

Traceback (most recent call last):
  File "/home/developer/.config/JetBrains/PyCharmCE2020.1/scratches/scratch_1.py", line 7, in <module>
    df = df.replace({"replace_string": "test"})
  File "/home/developer/PycharmProjects/pandas/pandas/core/frame.py", line 4277, in replace
    return super().replace(
  File "/home/developer/PycharmProjects/pandas/pandas/core/generic.py", line 6598, in replace
    return self.replace(
  File "/home/developer/PycharmProjects/pandas/pandas/core/frame.py", line 4277, in replace
    return super().replace(
  File "/home/developer/PycharmProjects/pandas/pandas/core/generic.py", line 6641, in replace
    new_data = self._mgr.replace_list(
  File "/home/developer/PycharmProjects/pandas/pandas/core/internals/managers.py", line 616, in replace_list
    masks = [comp(s, regex) for s in src_list]
  File "/home/developer/PycharmProjects/pandas/pandas/core/internals/managers.py", line 616, in <listcomp>
    masks = [comp(s, regex) for s in src_list]
  File "/home/developer/PycharmProjects/pandas/pandas/core/internals/managers.py", line 614, in comp
    return _compare_or_regex_search(values, s, regex)
  File "/home/developer/PycharmProjects/pandas/pandas/core/internals/managers.py", line 1946, in _compare_or_regex_search
    _check_comparison_types(False, a, b)
  File "/home/developer/PycharmProjects/pandas/pandas/core/internals/managers.py", line 1925, in _check_comparison_types
    raise TypeError(
TypeError: Cannot compare types 'ndarray(dtype=int64)' and 'str'

Process finished with exit code 1

Expected Output

The original DataFrame.

If at least one column is from dtype object, the replace works, for example the following works:

import pandas as pd

df = pd.DataFrame({"Y0": [1, "2"], "Y1": [3, 4]})
df = df.replace({"replace_string": "test"})

Unexpectedly the following Code works:

import pandas as pd

df = pd.DataFrame({"Y0": [1, 2], "Y1": [3, 4]})
df = df.replace(to_replace="replace_string", value="test")

This is probably related to #16784

Tested on master.