profiling: write high level profiling data of the extraction process

[vlaci]

To try the visualization without checking out this branch, here is a sample file (zip). Extract it, then upload it to https://speedscope.app

Questions:

• [x] Q: Where should we put the profile files? It makes sense to put them next to the extract folder or also into the extract folder. A: As there can be multiple input files, but we could write a single profile for the whole run, so it makes sense to expect the output location as an argument.

[vlaci]

There is still some tests to write. I thought about creating a sample list of events and assert on the contents of the generated json. Do you have better ideas?