Early launch crash about revision

[Yattabyte]

The crash appears to happen in the nvidia driver, coming from the main thread.
At the time of the crash, none of the other threads were doing anything with the main or secondary rendering context.

I've recently added more safeguards for reading/writing from multiple threads, but none have solved the issue.

[Yattabyte]

I've ruled out a few sources of error:

1. Sponza scene is still having its assimp importer object created
2. Same for ref_test
3. Material for test wall object still trying to retrieve pixel data with freeimage, can step through it
4. Same for grass data, can step through it

Going to have to hone in on the issue in the main thread slowly, since the call stack is all external code when the crash happens.

[Yattabyte]

I've painstakingly traced the main thread as far as it can go before crashing
It appears to crash some time after the 100th frame, when trying to render the bounding sphere for the first pointlight. I've traced this by outputting messages to the console before and after certain procedures, getting as far as glDrawArrays.

I don't think this crash happens when there's only a directional light in the scene.

In any case, the data being rendered must be corrupted or something.

[Yattabyte]

I've replaced the sphere model, and the issue seems to have gone away.
In any case, the model I was using was ridiculously high poly for the usage it was getting.

Issue fixed in commit 3280991