The tests are fine, your code is wrong, why would you need to hardcode the test with 144 as input otherwise?

Incorrect answer for n=144: 441 should equal 414

You can clearly see that the next bigger number with the digits of 144 is indeed 414, not 441 as your code (without the hardcoding part) returns.

When you think there is a problem with the tests, first check how many completions in the language you're using the kata already has:

Python Completions 11785

Then provide some proof about what you think is wrong: language, input, expected output, your output, optionally, your code.

This bug report is bad.

The test results are incorrect