implement LoRA / QLoRA (#46)

* implement LoRA / QLoRA

- example of using MLX to fine-tune an LLM with low rank adaptation (LoRA) for a target task
- see also https://arxiv.org/abs/2106.09685
- based on https://github.com/ml-explore/mlx-examples/tree/main/lora

* add some command line flags I found useful during use
- --quiet -- don't print decorator text, just the generated text
- --prompt @/tmp/file.txt -- load prompt from file

* user can specify path to model OR model identifier in huggingface

* update mlx-swift reference

Co-authored-by: Ashraful Islam <ashraful.meche@gmail.com>
Co-authored-by: JustinMeans <46542161+JustinMeans@users.noreply.github.com>

Libraries/LLM/Tokenizer.swift

@@ -4,10 +4,20 @@ import Foundation
 import Hub
 import Tokenizers
 
-public func loadTokenizer(configuration: ModelConfiguration) async throws -> Tokenizer {
+public func loadTokenizer(configuration: ModelConfiguration, hub: HubApi) async throws -> Tokenizer
+{
     // from AutoTokenizer.from() -- this lets us override parts of the configuration
-    let config = LanguageModelConfigurationFromHub(
-        modelName: configuration.tokenizerId ?? configuration.id)
+
+    let config: LanguageModelConfigurationFromHub
+
+    switch configuration.id {
+    case .id(let id):
+        config = LanguageModelConfigurationFromHub(
+            modelName: configuration.tokenizerId ?? id, hubApi: hub)
+    case .directory(let directory):
+        config = LanguageModelConfigurationFromHub(modelFolder: directory, hubApi: hub)
+    }
+
     guard var tokenizerConfig = try await config.tokenizerConfig else {
         throw LLMError(message: "missing config")
     }