Remove the **MAGIC**

Due to the goal of this library that mirrors Protractor which in turn justifiable claims that "[y]ou no longer need to add waits and sleeps to your test" and "you don’t have to worry about waiting for your test and webpage to sync." everything it accomplishes in this regard feels like MAGIC. This complicates the situation when things don't go as expected. It also downplays the importance of the end users understanding what is going on, what this library can and cannot accomplish.

This ticket lists some of the items I want to change to help remove the magic.

1. If the library has to poll (more than once, twice, N times) than an info log statement is recorded.
2. If the library timeouts log that piece of information at info level.
3. [.. something else that I really can't remember at this time ..]

未指定文件、测试或入口点。首先定位库中的 polling 和 timeout 路径，然后确定当前如何报告重复 polling 和 timeout。完成标准是：约定的 polling 和 timeout 事件以 info 级别记录日志，并明确剩余范围。