consolidate the tokenizer (#14327)

This change adds a new x/tokenizer package which includes:
  * New BPE and SentencePiece tokenizers
  * Removing the dependency on the imagegen tokenizers
  * Fixes to multibyte decoding in the pipeline
  * Various correctness and benchmark tests

Not included in this PR is the WordPiece tokenizer for BERT models which will be
added when we add embedding models. The imagegen tokenizers will also be removed in
a follow-up PR.

x/tokenizer/tokenizer_load_test.go

@@ -0,0 +1,26 @@
+//go:build mlx
+
+package tokenizer
+
+import (
+	"strings"
+	"testing"
+)
+
+func TestLoadFromBytesRejectsWordPiece(t *testing.T) {
+	data := []byte(`{
+		"model": {
+			"type": "WordPiece",
+			"vocab": {"[UNK]": 0, "hello": 1}
+		},
+		"added_tokens": []
+	}`)
+
+	_, err := LoadFromBytes(data)
+	if err == nil {
+		t.Fatal("expected WordPiece load to fail")
+	}
+	if !strings.Contains(err.Error(), "unsupported tokenizer type: WordPiece") {
+		t.Fatalf("unexpected error: %v", err)
+	}
+}