add new finalize_run step to iterate forecast process

[mitokic]

This pull request introduces enhancements to the agent forecasting workflow in the finnts package, focusing on better tracking and management of forecast run iterations and completion status. The changes improve how the system determines when to skip or proceed with model optimization, add metadata for backward compatibility, and introduce a new mechanism for finalizing agent runs. Additionally, minor version and documentation updates are included.

Agent run tracking and workflow improvements:

• Added max_iterations and run_complete columns to agent run tracking, allowing more precise control over when model optimization should be skipped or continued. This includes backward compatibility for existing runs and ensures local models only run when necessary. (R/agent_iterate_forecast.R, [1] [2]
• Introduced the finalize_run function, which updates metadata for each combo after a run completes, marking runs as complete and setting the max iteration count. This function is now integrated into the agent workflow, replacing the previous "stop" node with "finalize_run" for proper finalization. (R/agent_iterate_forecast.R, [1] [2] [3]

Data and code hygiene:

• Updated the saving process for best agent runs to remove unnecessary columns (run_complete, max_iterations) before persisting results, keeping log files clean. (R/agent_iterate_forecast.R, R/agent_iterate_forecast.RR794-R797)
• Registered new global variables (max_iterations, run_complete) to avoid warnings in package checks. (R/utility.R, R/utility.RL30-R31)

Documentation and versioning:

• Incremented package version to 0.6.0.9014 and updated the news file to reflect the latest development changes. (DESCRIPTION, [1]; NEWS.md, [2]…er when iterate process has to be restarted

[Copilot]

Pull request overview

This PR enhances the agent forecasting workflow by introducing a new finalize_run step that properly tracks forecast run completion and iteration counts. The changes enable better management of model optimization, ensuring local models only run when necessary and allowing interrupted processes to resume correctly.

Key Changes:

• Added max_iterations and run_complete columns to agent run tracking with backward compatibility support for existing runs
• Introduced a new finalize_run function integrated into the workflow to update metadata when runs complete
• Modified the workflow to route through finalize_run before stopping, replacing direct routing to the "stop" node

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 6 comments.

File	Description
R/agent_iterate_forecast.R	Implements finalize_run function, adds run completion tracking logic, updates workflow nodes to route through finalize_run, and adds backward compatibility for new metadata columns
R/utility.R	Registers new global variables (max_iterations, run_complete) to prevent package check warnings
NEWS.md	Updates version number to 0.6.0.9014 for development tracking
DESCRIPTION	Bumps package version to 0.6.0.9014

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.