perf: Replace LCOV setup with native AWK parsing for coverage reports

[taorepoara]

Summary

Refactored the coverage report parsing workflow to eliminate the external LCOV setup dependency and instead parse coverage metrics directly using AWK, improving workflow efficiency and reducing external dependencies.

Key Changes

• Removed the setup-lcov action step that was previously required to parse coverage reports
• Replaced bash regex parsing with native AWK script that directly processes LCOV coverage file format
• AWK script extracts coverage metrics (lines, functions, branches) by parsing LF, LH, FNF, FNH, BRF, and BRH fields
• Calculates coverage percentages inline with proper handling of division-by-zero cases
• Maintains identical output format for downstream workflow steps

Implementation Details

• The new AWK approach reads the coverage file directly without requiring external tools
• Coverage percentages are calculated with one decimal place precision using printf "%.1f\n"
• Handles edge cases where total counts are zero to prevent division errors
• Simplifies the workflow by removing an external action dependency while maintaining the same functionality

https://claude.ai/code/session_019LbNU7m5v2JAzSMfRvEcGA