JAX FP8 matmul (fusion) Jupyter Notebook tutorial. (#133)

In this tutorial notebook, we investigate how the ML stack JAX + XLA handles the specificities of FP8 matmuls,
while still generating an optimal fused kernel call including:
* FP8 inputs scaling;
* FP8 output scaling & clamping;
* Non-linearity & bias fusing;
* Abs-max output capture;

Note: some open questions remain on bias or gelu fusing.

.gitignore

@@ -80,6 +80,7 @@ target/
 
 # Jupyter Notebook
 .ipynb_checkpoints
+.jupyter_ystore.db
 
 # IPython
 profile_default/