Improve goal execution timing precision by excluding scheduling overhead

[Copilot]

This PR addresses timing measurement accuracy in goal execution by separating scheduling overhead from actual function execution time. Previously, the timing measurement included overhead from function imports, transaction setup, and context preparation, making it difficult to get accurate performance metrics for goal handlers.

Problem

The existing timing measurement in _pursue_goal_core() captured the total time from the start of goal processing to completion, which included:

• import_string() overhead for loading the handler function
• Transaction setup and context preparation
• Thread local context management
• Argument extraction and preparation

This made it impossible to distinguish between scheduling/setup time and actual execution time, especially for goals that were initially blocked and later unblocked.

Solution

This PR moves timing measurement deeper into the execution flow to capture only the actual function execution time:

1. Precise timing in follow_instructions(): Timing now starts just before the actual handler function call and ends immediately after, excluding all setup overhead.

2. Thread-local timing storage: Added execution_time field to GoalsThreadLocal to pass precise timing from follow_instructions() to _pursue_goal_core().

3. Robust exception handling: Ensures timing is captured correctly even when exceptions occur during execution.

Example

def my_handler(goal):
    # Only this execution time is now measured
    time.sleep(0.1)  # Will show ~100ms, not 100ms + import/setup overhead
    return AllDone()

# Schedule as blocked for clean timing separation
goal = schedule(my_handler, blocked=True)

# Later unblock and execute - timing captures only handler execution
unblocked_goal = unblock_retry_goal(goal.id)

Benefits

• Accurate performance metrics: Goal execution timing no longer includes scheduling overhead
• Better debugging: Clear separation between scheduling time and execution time
• Improved monitoring: More precise performance measurements for optimization
• Backward compatibility: Existing blocked parameter behavior is preserved

The changes are minimal (27 lines added, 3 modified) and maintain full backward compatibility while providing significantly more accurate timing data for goal execution monitoring and optimization.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• esm.ubuntu.com
  • Triggering command: /usr/lib/apt/methods/https (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.