Prevent Azure Monitor metrics from leaking into consumer application logs

[ivan-nikalaichanka-4ss]

This PR is related to user story ESS-4314

Description

When Application Insights integration is enabled, the client_metric_appinsight logger emits messages that propagate up to the datareservoirio logger and beyond. Since ancestor loggers' levels and filters are not considered during propagation (only handler-level filtering applies), these internal telemetry messages leak into any handler a consuming application has attached to datareservoirio - regardless of the log level the application has configured.

Checklist

• PR title is descriptive and fit for injection into release notes (see tips below)
• Correct label(s) are used

[Copilot]

Pull request overview

This PR addresses a logging-propagation issue where internal Azure Monitor / Application Insights metric logs from the client metrics logger bubble up into consumer applications’ handlers attached to datareservoirio, even when consumers configure restrictive logger levels.

Changes:

• Disable logger propagation for the client metrics Application Insights logger when AI integration is enabled.
• Add inline documentation pointing to Python logging propagation behavior.

Suppressed comments (1)

datareservoirio/client.py:49

• This change is intended to stop the metric logger from propagating into consumer-installed handlers. There’s no unit test covering that metric().propagate becomes False when Application Insights is enabled, which makes regressions likely (especially since metric() is cached). Consider adding a test that sets ENV_VAR_ENABLE_APP_INSIGHTS, stubs _ensure_azure_monitor_configured/configure_azure_monitor, calls metric(), and asserts logger.propagate is False.

            logger.propagate = False
            logger.setLevel(logging.DEBUG)
            _ensure_azure_monitor_configured(
                connection_string=environment._application_insight_connectionstring,
                logger_name=__name__ + "_metric_appinsight",

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.