Correctly handle TypeOfAny.from_another_any

[ilevkivskyi]

Fixes #15253

The fix is trivial so I am not going to wait for a review, will just check the primer. Also I am trying something new here: adding a unit test.

[ilevkivskyi]

And of course this uncovered a big can of worms: currently TypeOfAny.special_form is used in a lot of places where it is a totally real Any but the source Any is not accessible, for example an attribute access on a class with fallback_to_any. Overload ambiguity uses has_any_type() and will actually randomly choose first matching overload for those "special form" Any, while it actually should count them as real.

The only reasonable solution I see here is to introduce a new kind of Any, and move some part of existing special_form Anys to it (and there are now ~100 uses). This new type will be counted as a real one (for both --disallow-any-expr and for overload ambiguity etc), but this is actually a big chunk of work, so I don't really want to do it.

[github-actions]

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

pandas (https://github.com/pandas-dev/pandas)
+ pandas/core/arrays/interval.py:1019: error: Incompatible types in assignment (expression has type "DatetimeArray | TimedeltaArray | ndarray[Any, Any]", variable has type "ndarray[Any, dtype[Any]]")  [assignment]
+ pandas/core/arrays/sparse/array.py:1879: error: Incompatible types in assignment (expression has type "ExtensionArray | ndarray[Any, Any]", variable has type "ndarray[Any, Any]")  [assignment]
+ pandas/core/window/rolling.py:1683: error: "bool" has no attribute "astype"  [attr-defined]
+ pandas/core/window/rolling.py:1730: error: "bool" has no attribute "astype"  [attr-defined]
+ pandas/core/groupby/groupby.py:5349: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[Any]]", variable has type "list[Any]")  [assignment]
+ pandas/core/groupby/generic.py:793: error: Incompatible types in assignment (expression has type "list[ndarray[Any, dtype[_SCT]]]", variable has type "list[ndarray[Any, dtype[signedinteger[Any]]]]")  [assignment]

pandas-stubs (https://github.com/pandas-dev/pandas-stubs)
- tests/test_pandas.py:245: error: Expression is of type "Any", not "Series[bool]"  [assert-type]
+ tests/test_pandas.py:245: error: Expression is of type "DataFrame", not "Series[bool]"  [assert-type]

optuna (https://github.com/optuna/optuna)
+ optuna/importance/_fanova/_tree.py:64: error: Incompatible types in assignment (expression has type "tuple[Any, ...]", variable has type "list[Any]")  [assignment]

schemathesis (https://github.com/schemathesis/schemathesis)
+ src/schemathesis/specs/openapi/schemas.py: note: In member "get_operation_by_reference" of class "BaseOpenAPISchema":
+ src/schemathesis/specs/openapi/schemas.py:315: error: Argument 2 to "chain" has incompatible type "Dict[str, Any]"; expected "Iterable[Dict[str, Any]]"  [arg-type]

scrapy (https://github.com/scrapy/scrapy)
+ scrapy/http/request/form.py:52: error: Argument 1 to "_set_url" of "Request" has incompatible type "Literal[b'']"; expected "str"  [arg-type]