brpc pthread绑核

看brpc的代码，brpc负责运行bthread的内核线程(内核线程数默认和服务器的cpu逻辑核数相同)，没有做绑核操作，这是基于什么考虑啊？绑核性能是不是会更好？我们seastar的一些框架都会做绑核操作。

No file, test, or entry point is named. Start by tracing how brpc creates and schedules bthread kernel threads, then compare that behavior with the seastar pinning approach; done would require an agreed design and evidence about whether CPU affinity improves performance without harming brpc's scheduling model.

Written by the indexing model from the issue text.