Non utf8 python file test

[seddonym]

Adds a regression test for reading the contents of Python files that aren't UTF-8-encoded.

This file was created in the Python repl using:

>>> with open(filename, "w", encoding="shift_jis") as f:
...     f.write("# -*- coding: shift_jis -*-\n\nfrom .imported import π")

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #230 will not alter performance

_{Comparing non-utf8-python-file-test (7e4d506) with master (0941b8f)}

Summary

✅ 22 untouched benchmarks