outsource reconstruction data in separate arrays

Previously, all fit information data (corresponding key: 'fitinf') from offline files was given to the user in a big chuncked array. To address the issue #16 (closed) 'fitinf data' is restructured into a numpy rec array.

km3io/offline.py

@@ -537,14 +537,13 @@ class OfflineReader:
 
             yield trk_index, stages_index
 
-    def _find_empty(self, chunk_arr):
-        """finds empty lists/arrays in a chunked-jagged array of data
-        of interest.
+    def _find_empty(self, array):
+        """finds empty lists/arrays in an awkward array
 
         Parameters
         ----------
-        chunk_arr : Chunked array
-            Chunked array or jagged array of data of interest. For example:
+        array : awkward array
+            Awkward array of data of interest. For example:
             km3io.OfflineReader(my_file).tracks.fitinf .
 
         Yields
@@ -556,7 +555,7 @@ class OfflineReader:
             empty and does not contain an empty list, then False is written in the
             corresponding index.
         """
-        for i, rs in enumerate(chunk_arr):
+        for i, rs in enumerate(array):
             try:
                 if len(rs)==0:
                     j = None