MON-199535 Factorize Tools Implementation

[gregory-lvtx]

Ticket Jira: MON-199535

What ?

Factorize tool implementations.

Why ?

Some components related to models inheriting from the same mixins have a common implementation of tools such as create, list, and delete. Defining generic CRUD tools will simplify the code by removing redundancy.

How ?

• Define generic _create, _list, _delete, _patch methods in a new components.base.py.
• Update each relevant component with these new generic methods.
• Update unit tests to align with the new generic methods.

[denrou]

🔍 Automated Review — Action Required

Reviewed with the Centreon automated review skill. Complexity: medium–high — Recommended reviewers: 2. All 84 tests pass and CI is green; the refactor itself is clean and consistent. See inline comments for details.

🔴 Critical

• Default sort order dropped for all list endpoints — components/base.py. The new _list sends sort_by=None when no order is passed, where the old helper always applied each model's default sort field. This silently changes default result ordering (and pagination) for all 11 list tools. See inline comment.

🟠 Important

• No test coverage for the new generic helpers — the component tests mock _list/_create/_delete/_patch, so components/base.py is never executed (regression for _delete's aggregation logic). See inline comment.
• Dead code: the old generic _list in utils/base.py (~L37) is now unused after this refactor and can be removed.

🟡 Suggestion

• Stale test comment in test_component_host.py. See inline comment.

⚠️ The Critical item is a silent behavior change — please confirm whether it's intentional before merge.

🤖 Generated with Claude Code

[denrou]

These new generic helpers have no direct test coverage — every component test mocks _list/_create/_delete/_patch at the module level, so the real implementations here are never executed. This is a coverage regression for _delete in particular: the asyncio.gather(..., return_exceptions=True) + dict(zip(..., strict=True)) aggregation previously ran for real in each component delete test (only Model.delete was mocked). A dedicated test_component_base.py covering _delete success/exception paths and _list's sort_by would also catch the default-ordering issue above.