sampling : avoid expensive softmax during greedy sampling (#9605)

* sampling : avoid expensive softmax during greedy sampling

ggml-ci

* speculative : fix default RNG seed + set sparams.n_probs

* Update tests/test-sampling.cpp

Co-authored-by: slaren <slarengh@gmail.com>

* sampling : add clarifying comment [no ci]

---------

Co-authored-by: slaren <slarengh@gmail.com>

src/llama-sampling.cpp

@@ -3,13 +3,14 @@
 #include "llama-vocab.h"
 #include "llama-grammar.h"
 
-#include <cassert>
 #include <algorithm>
-#include <cstring>
-#include <ctime>
+#include <cassert>
 #include <cfloat>
 #include <chrono>
 #include <cmath>
+#include <cstdlib>
+#include <cstring>
+#include <ctime>
 #include <numeric>
 #include <random>
 #include <unordered_map>