test: add coverage for OllamaLLM edge cases and tool conversion

[goblin0xcrypto]

🧪 Improve Test Coverage for OllamaLLM

Summary

This PR adds additional test cases to improve code coverage for the OllamaLLM class, focusing on edge cases and previously untested functionality.

Changes

• ✅ Add test for tool schema conversion with actual function schemas
• ✅ Add test for JSON parsing error handling
• ✅ Add test for base URL with trailing slash normalization
• ✅ Add test for tool calls with string-formatted arguments

Test Cases Added

1. test_convert_tools_with_schemas

   • Validates tool conversion logic when function schemas are present
   • Ensures proper Ollama format structure

2. test_ask_json_parse_error

   • Tests resilience when API returns invalid JSON
   • Verifies graceful error handling

3. test_base_url_trailing_slash

   • Confirms URL normalization removes trailing slashes
   • Ensures correct endpoint construction

4. test_tool_calls_with_string_arguments

   • Tests handling of tool call arguments as JSON strings
   • Validates proper parsing and action conversion

Testing

All tests pass successfully:

pytest tests/test_ollama_llm.py -v

Coverage Impact

These additions increase test coverage for:

• _convert_tools_to_ollama_format method
• Error handling paths in ask method
• Configuration initialization edge cases
• Tool call argument parsing variations

[goblin0xcrypto]

Hi @openminddev
It looks like the required checks are still pending because the workflows are awaiting approval.
Would you mind approving the workflows so the CI can run?
Thanks a lot!

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ All tests successful. No failed tests found.

📢 Thoughts on this report? Let us know!