Conformance - fix issues on Windows (unicode, Pyre)

[Andrej730]

1. Noticed that pyright is giving unicode output that on Windows is leading to issues generating reports later. Saving this output explicitly as unicode resolves the issue.
2. Made Pyre optional as it does not support Windows currently (pyre doesn't work under windows facebook/pyre-check#554)
    
   Unicode error traceback:

Generating summary report
Traceback (most recent call last):
  File "\typing\conformance\src\main.py", line 260, in <module>
    main()
  File "\typing\conformance\src\main.py", line 256, in main
    generate_summary(root_dir)
  File "\typing\conformance\src\reporting.py", line 19, in generate_summary
    summary = template.replace("{{summary}}", generate_summary_html(root_dir))
                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "\typing\conformance\src\reporting.py", line 87, in generate_summary_html
    results = tomli.load(f)
              ^^^^^^^^^^^^^
  File "src\tomli\_parser.py", line 134, in load
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xa0 in position 110: invalid start byte