Showing benchmarks in README ?

Hi, I came across pyjson5 mainly because the [json5](https://pypi.org/project/json5) project was very clear that it is slow.

My initial reaction was pyjson5 would be even slower because that's normally how libraries starting with py are :D (For example: pysqlite vs sqlite)
But when I saw this was actually written in cython, I was interested to know if it is faster than `json5`.

My very small microbenchmark shows:
```
# Large file: 2000x faster 🚀
In [4]: %timeit json5.loads(pathlib.Path('test_400kb.txt').read_text())
9.33 s ± 696 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

In [5]: %timeit pyjson5.loads(pathlib.Path('test_400kb.txt').read_text())
4.38 ms ± 206 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)

# Small file: 1300x faster 🚀
In [6]: %timeit json5.loads(pathlib.Path('test_20kb.txt').read_text())
174 ms ± 9.42 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)

In [7]: %timeit pyjson5.loads(pathlib.Path('test_20kb.txt').read_text())
126 μs ± 7.3 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)
```

And that was a really amazing speed up.

I think it might be a good idea to highlight that in the README with some good benchmarks. For example: https://github.com/spyoungtech/json-five-rs#benchmarking

Start with the README and the benchmark example linked in the issue. Reproduce the json5 and pyjson5 loads comparisons using test_400kb.txt and test_20kb.txt, then document a clear benchmark summary and methodology in the README; done when readers can see the performance comparison and how it was measured.

Written by the indexing model from the issue text.