Add support for multiple streams and metadata parsing to 'ffmpeg_parse_infos'

[mondeja]

• Added support for metadata parsing to video.io.ffmpeg_reader.ffmpeg_parse_infos function.
• Added support for multiple streams parsing to video.io.ffmpeg_reader.ffmpeg_parse_infos function (first step to fix in the future multiple input audio streams not getting merged #178).
• Added support for chapters parsing to video.io.ffmpeg_reader.ffmpeg_parse_infos function.
• Added -hide_banner option to ffmpeg command in ffmpeg_parse_infos to remove the ffmpeg banner for better performance.
• Increased parsing speed of ffmpeg_parse_infos, all the parsing is done in a single loop. After chapters parsing added, the speed of both implementations is the same.
• "video_rotation" field returned by ffmpeg_parse_infos is placed now at ["video_metadata"]["rotate"].
• Removed test_ffmpeg_parse_infos_for_i926 because the same behaviour is checked in another test.

---

• I have provided code that clearly demonstrates the bug and that only works correctly when applying this fix
• I have added suitable tests demonstrating a fixed bug or new/changed feature to the test suite in tests/
• I have properly documented new or changed features in the documentation or in the docstrings
• I have properly explained unusual or unexpected code in the comments around it
• I have formatted my code using black -t py36

[coveralls]

Coverage increased (+0.7%) to 66.76% when pulling 8aeed0c on mondeja:ffmpeg_parse_infos-metadata into a378771 on Zulko:master.

[mondeja]

I think this implementation is stable enough now, so I go ahead with the merge...