Change input files to use tab-separated files instead of white-space separated files

[Las02]

Currently, the pipeline will not work if there is whitespace in the path, but this is common for people to do. Lots of people have directory names like reads for plasmaag . Instead, make it explicitly a tab-separated file, and then

• Document this on the README
• Re-format the example file so it's clear it uses tabs
• When reading the file, make sure it fails if any of the paths contain a tab
• The test data for the input contains a config file on zenodo using whitespace, which needs to be changed
• The help options shows examples using whitespace which needs to be changed
  Suggested by @jakobnissen.