Prevent context-less arg inference causing messages

[A5rocks]

I discovered this while investigating #20013. I'm not sure there's a better test case though. This PR also masks some errors (by providing a better type context for some messages), but I feel like that's alright.

This PR ensures that we don't output messages from the first pass of overloads, where we decide what to do.

[A5rocks]

Primer seems not great, I'll have to take a better look later...

[github-actions]

Diff from mypy_primer, showing the effect of this PR on open source code:

cki-lib (https://gitlab.com/cki-project/cki-lib)
- tests/kcidb/test_validate.py:200: error: Unpacked dict entry 0 has incompatible type "Collection[str]"; expected "SupportsKeysAndGetItem[str, str]"  [dict-item]

sphinx (https://github.com/sphinx-doc/sphinx)
- sphinx/transforms/__init__.py: note: In member "apply" of class "GlossarySorter":
+ sphinx/transforms/__init__.py: note: In function "apply":

Expression (https://github.com/cognitedata/Expression)
- expression/extra/parser.py:276: error: Argument 1 to "join" of "str" has incompatible type "_A"; expected "Iterable[str]"  [arg-type]
- README.py:287: error: Unsupported operand types for < ("int" and "_TSource")  [operator]
- README.py:288: error: Unsupported operand types for + ("int" and "_TSource")  [operator]

freqtrade (https://github.com/freqtrade/freqtrade)
- freqtrade/templates/FreqaiExampleStrategy.py:245: error: Unsupported left operand type for & ("tuple[Index[Any] | Series[builtins.bool] | ndarray[tuple[Any, ...], dtype[numpy.bool[builtins.bool]]] | list[builtins.bool] | str | bytes | date | datetime | timedelta | <7 more items> | complex | integer[Any] | floating[Any] | complexfloating[Any, Any] | list[Any] | slice[Any, Any, Any] | tuple[str | bytes | date | datetime | timedelta | <7 more items> | complex | integer[Any] | floating[Any] | complexfloating[Any, Any], ...], ...]")  [operator]
- freqtrade/templates/FreqaiExampleStrategy.py:255: error: Unsupported left operand type for & ("tuple[Index[Any] | Series[builtins.bool] | ndarray[tuple[Any, ...], dtype[numpy.bool[builtins.bool]]] | list[builtins.bool] | str | bytes | date | datetime | timedelta | <7 more items> | complex | integer[Any] | floating[Any] | complexfloating[Any, Any] | list[Any] | slice[Any, Any, Any] | tuple[str | bytes | date | datetime | timedelta | <7 more items> | complex | integer[Any] | floating[Any] | complexfloating[Any, Any], ...], ...]")  [operator]