Merge pull request #307 from toohamster/master

fixed EdgeNgramTokenizer split word count

src/Support/EdgeNgramTokenizer.php

@@ -13,7 +13,7 @@ public function tokenize($text, $stopwords = [])
         $splits = preg_split($this->getPattern(), $text, -1, PREG_SPLIT_NO_EMPTY);
 
         foreach ($splits as $split) {
-            for ($i = 2; $i <= strlen($split); $i++) {
+            for ($i = 2; $i <= mb_strlen($split); $i++) {
                 $ngrams[] = mb_substr($split, 0, $i);
             }
         }