<fix>(para_split_v2): index out of range issue of span_text first char (#396)

Co-authored-by: liukaiwen <liukaiwen@pjlab.org.cn>

magic_pdf/para/para_split_v2.py

@@ -100,7 +100,7 @@ def __detect_list_lines(lines, new_layout_bboxes, lang):
 
     if lang != 'en':
         return lines, None
-    else:
+
     total_lines = len(lines)
     line_fea_encode = []
     """
@@ -114,6 +114,9 @@ def __detect_list_lines(lines, new_layout_bboxes, lang):
         x_map_tag_dict, min_x_tag = cluster_line_x(lines)
     for l in lines:
         span_text = __get_span_text(l['spans'][0])
+        if not span_text:
+            line_fea_encode.append(0)
+            continue
         first_char = span_text[0]
         layout = __find_layout_bbox_by_line(l['bbox'], new_layout_bboxes)
         if not layout: