You are given an integer, requests which denotes the number of requests, and a list wait_time, where wait_time[i] denote the maximum wait time of the i'th request. The requests in the list are served from the first index to the last, and each request will be served for one second, then it is removed from the list (essentially, when you are at index 0, your current time is 0 seconds, then you spend 1 second serving request 0 and remove it from the list). Meanwhile, if some requests in the rest of the list to the right of the current request exceeds its wait time, then it is removed from the list as well. Return an array denoting how many active requests are present in the list each second.