Add eleven initial corpora and new seeds

[murchandamus]

Adds new seeds per same methodology as in #186 and adds initial corpora for the following eleven fuzz targets:

• clusterlin_add_dependency
• clusterlin_ancestor_finder
• clusterlin_chunking
• clusterlin_cluster_serialization
• clusterlin_depgraph_serialization
• clusterlin_linearization_chunking
• clusterlin_linearize
• crypter
• crypto_aeadchacha20poly1305
• crypto_fschacha20poly1305
• p2p_handshake

[murchandamus]

The number of lines added here seems off. Looking into what happened there.

[maflcko]

Would be nice to add p2p_handshake as well. Either here or in a follow-up. (I'll do it later this month, when I submit my fuzz inputs, if no one beats me to it)

[murchandamus]

I checked and my version of Bitcoin Core did not have that fuzz target, then realized that it just got merged yesterday. I can add it tomorrow to this PR after fuzzing it the next day or so.

[murchandamus]

Still around 240,000 new seeds, so I guess I didn’t do it wrong the first time after all.

[dergoegge]

Still around 240,000 new seeds, so I guess I didn’t do it wrong the first time after all.

$ git diff --diff-filter=A --name-only origin/main..c98fc19a89d6c54a4b92d8e51a9b49d536d02ed2 | wc -l
1496
$ git diff --diff-filter=A --name-only origin/main..murch/2024-08-add-seeds | wc -l
19535

Only 1.5k new seeds for the new harnesses and about 20k new seeds in total, which seems fine.

I think the ~240k number refers to added lines (not sure how much sense that makes w.r.t fuzz inputs).

[murchandamus]

I see. I was assuming that any fuzz seed would be a single line

[dergoegge]

lgtm, waiting on CI

[maflcko]

You'll have to revert the changes to that folder, or otherwise exclude the inputs.

See bitcoin/bitcoin#30514

Maybe a temporary revert for just that folder, so that they can be added back afterwards easily?

[murchandamus]

I have reverted the addition of seeds for the utxo_snapshot target

[maflcko]

I've added a reminder to bitcoin/bitcoin#30514 (comment)