Switch back to subprocessing for llama.cpp

This should resolve a number of memory leak and stability defects by allowing
us to isolate llama.cpp in a separate process and shutdown when idle, and
gracefully restart if it has problems.  This also serves as a first step to be
able to run multiple copies to support multiple models concurrently.

.gitignore

@@ -10,4 +10,5 @@ ggml-metal.metal
 *.exe
 .idea
 test_data
-*.crt
+*.crt
+llm/build