[pdq][mlh] Add querying benchmark to /bin

[Dcallies]

Similar to #1696 , we want to add a tool to help us benchmark query lookup speeds.

The script should by default randomly generate PDQ hashes from noise.

At a minimum it should contains:

1. -q: The number of queries to run (default 1000 or something)
2. -b: The number of PDQ hashes to query against (default 10000 or so)
3. -d: What distance to use as a match (default 31)
4. --seed: A seed for the random number generator
5. -m: Method (default to brute force linear scan). This is the only method we will start with.

It should output:

1. Total time querying (if -b is very large, it can take up significant time, so make sure to only start the timer afterwards)
2. Time per query

---

In followups, we will add more methods like faiss, faiss_gpu, etc.