Optimise the way tracemalloc and PyRefTracer hooks work

In https://github.com/python/cpython/issues/125703 @markshannon has raised that he is unhappy about the performance implications of where these hooks are placed and in a call we discussed that he has some ideas on how to make them more performant by moving them elsewhere or adapting then.

I am opening this issue to track and sync about these improvements for 3.14 and beyond.

没有指定文件或测试。首先阅读 issue 125703 以及关于 tracemalloc 和 PyRefTracer hook 放置位置的相关讨论，然后检查相关的 CPython 运行时代码和性能测量结果。这里没有定义完成的标准；在开始实现之前，该 issue 需要一个达成共识的方法和一个可测量的 3.14 性能目标。