Add type hints.

This fixes the two errors:

```
axelrod/fingerprint.py:444: error: Function "numpy.array" is not valid as a type
22
axelrod/fingerprint.py:444: note: Perhaps you need "Callable[...]" or a callback protocol?
23
Found 1 error in 1 file (checked 1 source file)
```

and:

```
axelrod/strategies/ann.py:125: error: Incompatible types in assignment (expression has type "ndarray", variable has type "List[int]")
34
Found 1 error in 1 file (checked 1 source file)
```

note that in `fingerprint.py` I'm using the `ArrayType` type but that
when I tried to use that in `ann.py` a lot of other things broke.

Author: drvinceknight

axelrod/fingerprint.py

@@ -16,6 +16,8 @@
 from axelrod.strategy_transformers import DualTransformer, JossAnnTransformer
 from mpl_toolkits.axes_grid1 import make_axes_locatable
 
+from numpy.typing import ArrayLike
+
 Point = namedtuple("Point", "x y")
 
 
@@ -188,7 +190,7 @@ def _generate_data(interactions: dict, points: list, edges: list) -> dict:
     return point_scores
 
 
-def _reshape_data(data: dict, points: list, size: int) -> np.ndarray:
+def _reshape_data(data: dict, points: list, size: int) -> ArrayLike:
     """Shape the data so that it can be plotted easily.
 
     Parameters
@@ -441,7 +443,7 @@ def fingerprint(
         filename: str = None,
         progress_bar: bool = True,
         seed: int = None,
-    ) -> np.array:
+    ) -> ArrayLike:
         """Creates a spatial tournament to run the necessary matches to obtain
         fingerprint data.
 

axelrod/strategies/ann.py

@@ -22,7 +22,7 @@ def num_weights(num_features, num_hidden):
     return size
 
 
-def compute_features(player: Player, opponent: Player) -> List[int]:
+def compute_features(player: Player, opponent: Player) -> np.ndarray:
     """
     Compute history features for Neural Network:
     * Opponent's first move is C
@@ -91,7 +91,7 @@ def compute_features(player: Player, opponent: Player) -> List[int]:
     total_player_c = player.cooperations
     total_player_d = player.defections
 
-    return [
+    return np.array((
         opponent_first_c,
         opponent_first_d,
         opponent_second_c,
@@ -109,20 +109,19 @@ def compute_features(player: Player, opponent: Player) -> List[int]:
         total_player_c,
         total_player_d,
         len(player.history),
-    ]
+        ))
 
 
 def activate(
     bias: List[float],
     hidden: List[float],
     output: List[float],
-    inputs: List[int],
+    inputs: np.ndarray,
 ) -> float:
     """
     Compute the output of the neural network:
         output = relu(inputs * hidden_weights + bias) * output_weights
     """
-    inputs = np.array(inputs)
     hidden_values = bias + np.dot(hidden, inputs)
     hidden_values = relu(hidden_values)
     output_value = np.dot(hidden_values, output)