Fix/Add ReasoningResult confidence_tier/is_actionable computed fields + boundary tests

[ArokyaMatthew]

Fixes #117

What changed

• Added two computed fields on ReasoningResult:
  • confidence_tier: "high" (>= 0.75), "medium" (>= 0.50), otherwise "low"
  • is_actionable: True only when label == "Suspicious" and confidence >= 0.65
• Implemented them using Pydantic v2 @computed_field so they are included in model_dump(mode="json").
• Added tests/test_reasoning_schema.py with boundary + serialization coverage.

Tests

• Added unit tests covering tier boundaries: 0.49, 0.50, 0.74, 0.75, 1.0
• Added tests for is_actionable threshold + label behavior
• Verified both fields are serialized in model_dump(mode="json")

Summary by CodeRabbit

• Refactor

  • Enhanced reasoning result schema to properly expose computed fields for confidence assessment in API responses

• Tests

  • Added comprehensive test coverage for reasoning schema field computation, confidence tier boundaries, and serialization validation

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 6ab89b39-c807-4ffa-a16f-604e2dec0091

📥 Commits

Reviewing files that changed from the base of the PR and between eadd39f and 7d563e3.

📒 Files selected for processing (2)

• libs/schemas/reasoning.py
• tests/test_reasoning_schema.py

---

📝 Walkthrough

Walkthrough

This PR refactors ReasoningResult to use Pydantic's @computed_field decorator for exposing confidence_tier and is_actionable derived properties. The custom model_dump_json() override is removed, and comprehensive unit tests validate the threshold logic and JSON serialization behavior.

Changes

Computed fields implementation and testing

Layer / File(s)	Summary
Computed fields implementation libs/schemas/reasoning.py	Pydantic imports updated to include computed_field. ReasoningResult adds two computed properties: confidence_tier (derived from confidence thresholds: ≥0.75 → high, ≥0.50 → medium, otherwise low) and is_actionable (true when label=="Suspicious" AND confidence≥0.65). Custom JSON serialization method removed.
Computed field unit tests tests/test_reasoning_schema.py	Three parameterized pytest tests cover confidence-tier boundary values (0.49, 0.50, 0.74, 0.75, 1.0), is_actionable across label and confidence combinations, and verification that model_dump(mode="json") includes both computed fields with expected values.

🎯 2 (Simple) | ⏱️ ~10 minutes

🐰 Confidence tiers now computed, not hard-coded,
Actionability logic flows through schemas encoded,
Pydantic's decorator makes serialization clean,
Dashboard dashboards dashingly styled, pristine! 🎨✨

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and specifically summarizes the main changes: adding computed fields (confidence_tier and is_actionable) to ReasoningResult and adding boundary tests.
Linked Issues check	✅ Passed	The PR implements all requirements from issue #117: confidence_tier with correct thresholds (0.75, 0.50), is_actionable property with label and confidence checks, Pydantic @computed_field implementation for serialization, and comprehensive tests covering boundary values and serialization.
Out of Scope Changes check	✅ Passed	All changes are directly scoped to issue #117 requirements: modifications to libs/schemas/reasoning.py for computed fields and new test file tests/test_reasoning_schema.py with comprehensive tests.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[ArokyaMatthew]

@Devnil434 , please review

[kunal-9090]

I checked both failing CI jobs on this PR. Phase 3 failed because services/init.py eagerly imports tracking, which pulls unrelated tracking imports into memory tests. Phase 4 failed on Ruff errors in the reasoning package.\n\nI prepared a fix branch here: https://github.com/kunal-9090/Eagle/tree/codex/fix-pr-118-ci\n\nFix included:\n- stop eager tracking imports in services/init.py\n- correct services/tracking/tracker.py to import settings from libs.config.settings\n- clean Ruff issues in services/reasoning/vlm.py, services/reasoning/pipeline.py, and services/reasoning/scene_graph.py\n\nLocal verification passes for both lint commands:\n- python -m ruff check services/memory/ libs/schemas/memory.py services/init.py services/tracking/tracker.py\n- python -m ruff check services/reasoning/ libs/schemas/reasoning.py libs/schemas/memory.py services/init.py services/tracking/tracker.py\n\nLocal pytest is blocked only because this machine lacks akeredis, which the GitHub workflow installs. I could not push directly to this contributor fork; GitHub returned 403 for kunal-9090.