Fused FFN_UP+FFN_GATE op (#741)

* Fused up+gate+unary for regular (not MoE) FFN - CPU

* WIP CUDA

* Seems to be working on CUDA

For a dense model we get 2-3% speedup for PP and ~0.6% for TG.

* Add command line option

This time the option is ON by default, and one needs to turn it
off via -no-fug or --no-fused-up-gate

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>

common/common.h

@@ -191,6 +191,7 @@ struct gpt_params {
     int  mla_attn          = 0;     // MLA 0: standard attention, 1: MLA with K and transposed V cache, 2: MLA with just K cache
     int  attn_max_batch    = 0;     // Max batch size to use when computing attention (only applicable if flash_attn = false)
     bool fused_moe_up_gate = false; // fused up*unary(gate) op for MoE models
+    bool fused_up_gate     = true;  // fused up*unary(gate) op
     int  min_experts       = -1;
     float thresh_experts   = 0;