feat: add structured logging fields (log_template, log_args) to GELF output

[soustruh]

Summary

Adds a custom logging filter (StructuredLoggingFilter) that preserves Python's raw message template and arguments as separate GELF fields, enabling pattern-based filtering and aggregation in Datadog.

Why This Matters

Even with AI agents that can crawl unstructured logs, structured logging provides critical value:

• Precision: Structured fields are deterministic and unambiguous vs semantic text parsing
• Performance: Pattern matching on indexed fields is orders of magnitude faster than full-text search at scale
• Aggregation: Enables reliable analytics like count by log_template or grouping by argument values
• Agent Efficiency: Reduces token usage and enables agents to build reliable monitoring rules
• Cross-System Consistency: Group same error patterns across distributed systems for root cause analysis

Changes

• Added StructuredLoggingFilter class before CommonInterface to extract record.msg and record.args before interpolation
• Modified set_gelf_logger to attach the filter to the GELF handler
• New fields are included via existing include_extra_fields=True mechanism
• Removed stale # noqa: E301 comment from set_gelf_logger signature

Example

logging.warning("The Answer to the Ultimate Question of Life, the Universe, and Everything: %i", 42)

Results in GELF output with:

• short_message: "The Answer to the Ultimate Question of Life, the Universe, and Everything: 42"
• log_template: "The Answer to the Ultimate Question of Life, the Universe, and Everything: %i"
• log_args: [42]

Backward Compatibility

This is an additive change — existing log fields remain unchanged. Components adopting the new version will have additional structured data available for advanced filtering in Datadog.

Review & Testing Checklist for Human

• Verify StructuredLoggingFilter correctly extracts record.msg as template and record.args as list
• Confirm that when record.args is None (no arguments), log_args is set to None and doesn't break GELF serialization
• Test with a real GELF endpoint (or stdout=True) to verify log_template and log_args appear as extra fields in the output

Notes

• The filter is only attached to the GELF handler (not the default stdout logger), so local development logging is unaffected
• log_template and log_args are surfaced as GELF extra fields because include_extra_fields=True is the default

Link to Devin session: https://app.devin.ai/sessions/6eb5ae7e18be4f48ac4a1070c2fead88
Requested by: @soustruh

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring