BUG: Use np.integer instead of int, to allow unsigned integers

Additionally, make the fixes needed to avoid an incoming deprecation in numpy/numpy#9505

Author: eric-wieser

lib/matplotlib/mlab.py

@@ -3136,17 +3136,17 @@ def get_type(item, atype=int):
     def get_justify(colname, column, precision):
         ntype = column.dtype
 
-        if np.issubdtype(ntype, str) or np.issubdtype(ntype, bytes):
+        if np.issubdtype(ntype, np.character):
             fixed_width = int(ntype.str[2:])
             length = max(len(colname), fixed_width)
             return 0, length+padding, "%s"  # left justify
 
-        if np.issubdtype(ntype, int):
+        if np.issubdtype(ntype, np.integer):
             length = max(len(colname),
                          np.max(list(map(len, list(map(str, column))))))
             return 1, length+padding, "%d"  # right justify
 
-        if np.issubdtype(ntype, float):
+        if np.issubdtype(ntype, np.floating):
             fmt = "%." + str(precision) + "f"
             length = max(
                 len(colname),