add min_p sampling parameter (#2014)

Signed-off-by: Christopher Barrera <cb@arda.tx.rr.com>
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>

gpt4all-backend/llmodel.h

@@ -66,6 +66,7 @@ public:
         int32_t n_predict = 200;
         int32_t top_k = 40;
         float   top_p = 0.9f;
+        float   min_p = 0.0f;
         float   temp = 0.9f;
         int32_t n_batch = 9;
         float   repeat_penalty = 1.10f;