Python: honor reducer auto-reduce in chat completion threads

[pragnyanramtha]

Motivation and Context

Fixes #12303.

ChatHistoryAgentThread can be initialized with a ChatHistoryReducer, but new thread messages were appended with add_message(). That bypasses reducer add_message_async() behavior, so reducers configured with auto_reduce=True do not reduce history as messages are added through the agent thread.

Description

• call add_message_async() when the thread history is a ChatHistoryReducer
• keep the existing synchronous append path for plain ChatHistory
• add a regression test proving a truncation reducer with auto_reduce=True reduces newly added thread messages

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the SK Contribution Guidelines
• The code follows the .NET coding conventions and Python coding conventions
• All unit tests pass, and I have added new tests where possible
• I didn't break anyone 😄

Validation

From python/:

• uv run --python 3.12 pytest tests/unit/agents/chat_completion/test_chat_completion_agent.py -q -> 24 passed
• uv run --python 3.12 pytest tests/unit/contents/test_chat_history_truncation_reducer.py tests/unit/agents/chat_completion/test_chat_completion_agent.py -q -> 42 passed
• uv run ruff check semantic_kernel/agents/chat_completion/chat_completion_agent.py tests/unit/agents/chat_completion/test_chat_completion_agent.py -> passed
• git diff --check -> passed

[github-actions]

Automated Code Review

Reviewers: 4 | Confidence: 94% | Result: All clear

Reviewed: Correctness, Security Reliability, Test Coverage, Design Approach

---

Automated review by pragnyanramtha's agents

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[pragnyanramtha]

@microsoft-github-policy-service agree