Autograder is failing on correct function outputs

Hello, one of my functions are not passing the autograder, but I believe the output is correct:

```
📝 Guess flip with argument other than heads or tails

Usage: node guess-flip --call=[heads|tails]

❌ Guess flip with argument other than heads or tails
::error::The output for test Guess flip with argument other than heads or tails did not match%0AExpected:%0A.*Usage:\s?node\sguess-flip(.js)?\s--call=(\[|\(|\{)?\s?(heads?|tails?)\s?\|\s?["']?(heads?|tails?)["']?\s?(\]|\)\}).*%0AActual:%0A
```

For this test, the function seems to be behaving as it should, but the test is still failing. What am I doing wrong here?

Thanks!

Start with the `node guess-flip --call=[heads|tails]` entry point and the autograder case `Guess flip with argument other than heads or tails`. Compare the command's actual output with the expected usage pattern shown in the issue, then rerun that autograder case; done means the test accepts the output for an invalid argument.

Written by the indexing model from the issue text.