Overload evaluation does not conform to typing spec

Nobody has claimed this yet.

A recent addition to the typing spec conformance test uncovered a case that pyright is handling incorrectly.

This is found in the `overloads_evaluation.py` test in [this PR](https://github.com/python/typing/pull/1957).

```python
@overload
def example5(obj: list[int]) -> list[int]: ...

@overload
def example5(obj: list[str]) -> list[str]: ...

def example5(obj: Any) -> list[Any]:
return []

def check_example5(b: list[Any]) -> None:
assert_type(example5(b), Any)
```

Start with the overloads_evaluation.py conformance test referenced in the issue and inspect how pyright evaluates example5 with list[Any]. Compare the result with the typing specification and the linked typing conformance PR. The work is done when the test's assert_type(example5(b), Any) expectation passes.

Written by the indexing model from the issue text.