collisions VS sha1collisiondetection

That is still an attack on the second preimage or a collision resistance properties of the hash function. Most collisions do work this way, for example see [1].

[1] https://github.com/corkami/collisions

it's easy to do if you have a few hours to spare

remelem nem checksum ellenorzesre https://github.com/corkami/collisions !

Sure, here are two different screenshots of your comment made with this tool:

Hash collisions and exploitations – Instant MD5 collision\ (4 comments)

It's far more likely that somebody would try to artificially create a collision. There is code in git to try to detect this https://github.com/cr-marcstevens/sha1collisiondetection

First milestone is to be able to compile this code https://github.com/cr-marcstevens/sha1collisiondetection/ using Cesium.

For sha1, people made a system where you can detect the patterns that lead to a collision, and (for example) replace it with a different hash only for inputs that would be a problem. https://github.com/cr-marcstevens/sha1collisiondetection i think git does this to eek more life out of sha1.

I imagine you could take a similar counter-cryptnalysis approach to md5. (I am out of my depth here, so there could be reasons this doesnt work for md5 im unaware of)

I'm saying that Git is on my side of opinion, that's why Git added collision detection (hardened SHA1) and added support for more stronger hashes and Git's goal is transition out of SHA1. The same is with fossil, that added hardened SHA1 since 2017 and by default uses SHA3 for new repositories.