anton

first step was getting llama cpp to play nice with electron js so that we can run the model I fine tuned on the client, a couple errors but eventually got it wired up with node-llama-cpp bindings. this way the model + app can be shipped to the user (client) and run offline