The real reason to use a command line is that a command can go into a script, and a script runs without you.
That turns the slow, repetitive parts of running signage into something automatic. Provision a hundred screens in one run. Push a content change across a whole fleet on a schedule. Ship a custom app through a deployment pipeline, so a new version goes out the same way the rest of your code does, reviewed and released through the same process.
This is the same idea as the API, reached a different way. The API is for building Screenly into your own software; the CLI is for driving it from scripts and the terminal. Most teams use whichever fits the task in front of them.