Ensure everything works from a single script and produces a CSV

[alisonrclarke]

Ideal output: a big table/csv for each individual in each video, with all/selected markers.

MC: That would give us a complete pipeline from input video to output of CSV of movement data and visualisation for selected markers; parameters would include cropping the video (to focus on the people we want to track), optimising accuracy and smoothing.

[alisonrclarke]

Working on issue10_script_csv branch. Have created initial CLI script which outputs CSV file.

Still todo:

• [x] Write tests
• [x] Add docstrings and info to README
• [x] Add option to cli script to exclude body parts
• [x] Check CSV format with Martin and reshape as necessary
• [x] Add option for confidence threshold
• [x] Add option for jitter removal
• [x] Modify CoLab script to use method from the script
• [x] Ensure OpenPose output can be seen from CoLab script
• [x] Give option for saving CoLab output as a zip (either individually or all outputs)
• [x] Give option to upload JSON or video files?
• [x] Add parameter for number of people
• [x] Add option to pass additional flags to openpose