llama : remove n_threads from llama_decode_internal (#3614)

This commit removes `n_threads` from the `llama_decode_internal`
functions doc comment as it does not exist anymore.

It looks like this parameter was removed in
Commit 16bc66d ("llama.cpp : split
llama_context_params into model and context params").

Signed-off-by: Daniel Bevenius <[email protected]>

Author: danbev

llama.cpp

@@ -5721,7 +5721,6 @@ static struct ggml_cgraph * llama_build_graph(
 //
 //   - lctx:      llama context
 //   - batch:     batch to evaluate
-//   - n_threads: number of threads to use
 //
 // return 0 on success
 // return positive int on warning