[NA] [SDK] test: add unit test coverage for IsJson metric

[hasnain1241]

Details

Adds unit test coverage for the IsJson heuristic metric (sdks/python/src/opik/evaluation/metrics/heuristics/is_json.py), which previously had none. Covers valid JSON (object, array, and primitive types like number/string/boolean/null), invalid JSON (plain text, empty string, malformed JSON), and non-string input (None, int), documenting the metric's current behavior on each. is_json.py itself was not modified — this PR is test-only.

Change checklist

• User facing
• Documentation update

Issues

• Resolves [FR]: [SDK] Add missing unit test coverage for IsJson metric #7878
• OPIK-

AI-WATERMARK

AI-WATERMARK: yes

• Tools: Claude Code
• Model(s): Claude
• Scope: Drafted the test cases and test file structure, following existing conventions in test_heuristics.py / test_sentiment.py.
• Human verification: Reviewed the metric's source code and existing test patterns manually before requesting tests; ran the full test suite locally and verified all 11 tests pass before submission.

Testing

Ran the new test file locally in a virtual environment:
pytest sdks/python/tests/unit/evaluation/metrics/test_is_json.py -v

All 11 tests passed. Scenarios validated:

• Valid JSON: object, array, number, string, boolean, null → all score 1.0
• Invalid JSON: plain text, empty string, malformed JSON, None input, int input → all score 0.0

No regressions expected since is_json.py was not modified. Also ran make precommit (ruff, ruff-format, trailing whitespace, end-of-file checks) — all passed on the new file.

Documentation

N/A — this PR only adds test coverage, no user-facing behavior or docs changed.

[baz-reviewer]

Test bypasses public IsJson API

IsJson is imported from the internal opik.evaluation.metrics.heuristics.is_json, so this test can pass while the documented opik.evaluation.metrics export is broken — should we import IsJson from opik.evaluation.metrics instead, as .agents/skills/python-sdk/testing.md requires?

Want Baz to fix this for you? Activate Fixer You can also update your AI coding guidelines based on this comment by apply pr to [branch name]

Other fix methods

Prompt for AI Agents

Before applying, verify this suggestion against the current code. In
`sdks/python/tests/unit/evaluation/metrics/test_is_json.py` around line 3, update the
`IsJson` import used by these JSON metric tests to come from the public
`opik.evaluation.metrics` package rather than the internal `heuristics.is_json` module.
Keep the existing test cases unchanged so collection and execution verify that the
documented package-level export works correctly.

[hasnain1241]

I've reviewed the CLA (CLA.md) and agree to its terms for this contribution.