Is the content-length header can tell us something ?
It may depend on the particular server and application. In many cases, the HTTP response will contain the contents of the requested file itself, or the output of executing it. So usually, you'd just confirm that a file was returned.
If the content length for an invalid file is 0, and a valid file is > 0, then that could work. I don't know if it would be that cut and dry, however, but maybe you can identify a baseline value vs a valid file hit. But maybe instead the particular application instead returns an HTTP 500 code if the file is invalid. It's probably good to try files that you know exist, so you can calibrate the expected response based on how the server responds.
If RFI/LFI does not work on one (valid) file, does that mean there is no chance of this attack at all?
Depends how you define "valid". If the file exists but you do not have permission to read it, then you will probably get a negative response. There also may be path filtering or a forced file extension; in these cases you may only be able to view certain files.