Inline the per-item equality check in list.__contains__/index/count/remove #159092

Feature or enhancement

list.__contains__(), index(), count(), remove(), and element comparisons in list.__eq__() use PyObject_RichCompareBool() for each element.

During a scan, the search value is fixed. Since we already know its type and its value, when an element has the same exact built-in type (int, float, or str), its comparison implementation is already known. Going through generic rich-comparison dispatch adds unnecessary type-dispatch and slot-handling overhead.

Add a fast path for these exact-type pairs, fall back to PyObject_RichCompareBool() for other cases.

Has this already been discussed elsewhere?

This is a minor feature, which does not need previous discussion elsewhere

Links to previous discussion of this feature:

No response

Linked PRs

• gh-159092: Optimize comparisons in list operations #159093