Create a benchmark to measure reading/joining multiple files into a single DataFrame

It is a case where a user can read hundreds of files with the same structure and combine them into a single DataFrame later.
We need to explore the current solution with `concat` operation and compare with other possible approaches