llama : reuse compute graphs

ggml-ci

common/common.h

@@ -330,6 +330,7 @@ struct common_params {
     bool no_perf           = false; // disable performance metrics
     bool ctx_shift         = true;  // context shift on inifinite text generation
     bool swa_full          = false; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
+    bool graph_reuse       = false; // reuse previous compute graphs when possible
 
     bool input_prefix_bos  = false; // prefix BOS to user inputs, preceding input_prefix
     bool use_mmap          = true;  // use mmap for faster loads