Move gguf fuzzers to the llama.cpp repository

[slaren]

Fuzz testing of llama.cpp in OSS-Fuzz has been very valuable to detect leaks and security issues in the model loading code. Unfortunately, the build of the current fuzzers has been broken for a long time, and new code is not being tested.

We should move the fuzzers to this repository and ensure that they are maintained. More details: https://google.github.io/oss-fuzz/advanced-topics/ideal-integration/

@DavidKorczynski the current implementation seems to be Apache licensed, which would complicate moving the code here. Would it be possible to re-license it as MIT?

[DavidKorczynski]

Yeah that should be possible @slaren -- I can see to moving the harnesses over here, would that work?

[slaren]

Yes, that would be great. We can take over the maintenance of the tests.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.