[Hardik K (hardikkaurani)]

Description

This PR improves CompletionUsage in semantic_kernel.connectors.ai.completion_usage:

1. __radd__ and sum() support:
   Currently, attempting to aggregate multiple CompletionUsage objects using Python's standard sum([u1, u2, ...]) fails with:

   TypeError: unsupported operand type(s) for +: 'int' and 'CompletionUsage'

   This occurs because sum() starts with 0 as the default accumulator and delegates to __radd__. Implementing __radd__ allows sum() to work seamlessly on sequences of CompletionUsage objects.

2. Operand type validation in __add__:
   __add__ did not check operand types before accessing attributes like other.prompt_tokens. Adding an unsupported type previously raised an uninformative AttributeError: 'int' object has no attribute 'prompt_tokens' instead of returning NotImplemented to let Python raise the standard TypeError.

3. total_tokens property:
   Adds a computed total_tokens property ((self.prompt_tokens or 0) + (self.completion_tokens or 0)) matching provider conventions (e.g., openai.types.CompletionUsage).

Verification

• Added regression tests in python/tests/unit/connectors/ai/test_completion_token_usage.py covering:
  • sum([u1, u2, u3])
  • total_tokens computation on non-empty and empty instances
  • TypeError when adding unsupported operand types
• Ruff check and format pass cleanly.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

[Hardik K (hardikkaurani)]

Hi Eduard van Valkenburg (@eavanvalkenburg), could you please take a look at this PR when you have a moment? It adds radd support for sum(), validates operand types in add, and adds the total_tokens property on CompletionUsage. Tests and ruff checks pass cleanly. Thank you!