Add a comment indicating future work.

gpt4all-backend/CMakeLists.txt

@@ -101,6 +101,7 @@ foreach(BUILD_VARIANT IN LISTS BUILD_VARIANTS)
     prepare_target(replit-mainline llama-mainline)
 
     if (NOT LLAMA_METAL)
+# FIXME: These need to be forward ported to latest ggml
 #        add_library(gptj-${BUILD_VARIANT} SHARED
 #            gptj.cpp utils.h utils.cpp llmodel_shared.cpp llmodel_shared.h)
 #        prepare_target(gptj ggml-230511)
@@ -109,7 +110,7 @@ foreach(BUILD_VARIANT IN LISTS BUILD_VARIANTS)
             falcon.cpp utils.h utils.cpp llmodel_shared.cpp llmodel_shared.h)
         target_compile_definitions(falcon-${BUILD_VARIANT} PRIVATE LLAMA_VERSIONS=>=3 LLAMA_DATE=999999)
         prepare_target(falcon llama-mainline)
-
+# FIXME: These need to be forward ported to latest ggml
 #        add_library(mpt-${BUILD_VARIANT} SHARED
 #            mpt.cpp utils.h utils.cpp llmodel_shared.cpp llmodel_shared.h)
 #        prepare_target(mpt ggml-230511)