Refactor + Cleanup ScanContext

[Breakthrough]

The number of lines in the scan_motion() method of ScanContext is quite high, and lots of development artifacts still exist. Cleanup and reduce max function length by breaking existing method down into smaller parts, to improve reusability + eventual integration with new PySceneDetect API.

[Breakthrough]

Further complexity reduction can be achieved by re-using certain classes/modules from PySceneDetect, as well as reduce code duplication. In particular:

• Replace dvr_scan.platform with scenedetect.platform
• Replace FrameTimecode in dvr_scan.timecode with FrameTimecode from scenedetect
• Use scenedetect.VideoManager instead of manually handling video input/concatenation in ScanContext
• Use scenedetect.video_splitter to handle video output (would also resolve Extracted clips don't have audio #27, Motion file has bigger size than original #42)

The only downside to having scenedetect as a dependency is that it would bring in Click, which is not used by DVR-Scan. This may require splitting scenedetect into multiple packages, moving the CLI to an optional feature, or just transitioning DVR-Scan to use Click.

[Breakthrough]

Replacements with equivalents will not happen until v2.0. Right now focus should be on minimum changes to get basic testing functionality up and running, and close out development of v1.x right after to focus on 2.0.