llama: Add attention and final logit soft-capping, update scaling factor to Gemma2...

llama: Add attention and final logit soft-capping, update scaling factor to Gemma2 (#8197)

* Add attention and final logit softcapping.

* fix

* Add custom add_ functions

* Disable flash attention for Gemma2

* Update src/llama.cpp

Co-authored-by: slaren <redacted>
* Add default value for attention and final logit softcap value

* Add custom kq scaling from Gemma2Attention

* Remove custom pre attention scaling and use computed value instead.

---------

Co-authored-by: slaren <redacted>