Fix: Handle unhashable items in remove_duplicates

The previous optimization broke when the function received unhashable items
like lists or dicts, causing TypeError. This commit adds backward compatibility
by checking if items are hashable:
- Hashable items use set for O(1) lookup (fast path)
- Unhashable items fall back to list membership check (preserves original behavior)

This maintains the O(n) optimization for the common case while preserving
backward compatibility for all input types.

Co-Authored-By: Keon <[email protected]>

Author: devin-ai-integration[bot]

algorithms/arrays/remove_duplicates.py

@@ -7,17 +7,22 @@
 Input: [1, 1 ,1 ,2 ,2 ,3 ,4 ,4 ,"hey", "hey", "hello", True, True]
 Output: [1, 2, 3, 4, 'hey', 'hello']
 
-Time Complexity: O(n) where n is the length of the input array
+Time Complexity: O(n) for hashable items, O(n²) worst case for unhashable items
 Space Complexity: O(n) for the seen set and result array
 """
+from collections.abc import Hashable
 
 def remove_duplicates(array):
     seen = set()
     new_array = []
 
     for item in array:
-        if item not in seen:
-            seen.add(item)
-            new_array.append(item)
+        if isinstance(item, Hashable):
+            if item not in seen:
+                seen.add(item)
+                new_array.append(item)
+        else:
+            if item not in new_array:
+                new_array.append(item)
 
     return new_array