[Add] Gemma Example

[ariG23498]

In this PR I have added Gemma's example. I have also reported the results in the README.

[celunah]

this will also severely increase ram usage as your code upcasts BF16 into FP32

[hahmad2008]

What these params group_size_1 & group_size_2 represent?

    SE.Gemma.flash_self_extend_forward, group_size_1=8, group_size_2=1024

and how to get the desired extended context length based on the equation in the paper?
new_context_length = (old_context_length - neighbor_window) * group_size + neighbor_window