feat: Add standalone agent run mode inspired by LocalAGI

[localai-bot]

Add standalone agent run mode

This PR adds support for running agents in standalone mode, similar to LocalAGI's agent run command.

Features

• local-ai agent run <agent_name>: Run an agent by its ID from the registry
• local-ai agent run --config <file>: Run an agent from a JSON config file
• local-ai agent run <agent_name> --prompt "...": Foreground mode with single prompt
• Interactive mode: Continuous interaction with the agent

Implementation Details

• Added new AgentCMD struct in core/cli/agent.go
• Registered the agent subcommand in core/cli/cli.go
• Integrates with existing AgentPoolService for agent management
• Supports all existing agent pool configuration options

Usage Examples

# Run an agent by name
local-ai agent run my-agent

# Run an agent with a single prompt
local-ai agent run my-agent --prompt "Analyze this text"

# Run an agent from a config file
local-ai agent run --config agent-config.json

# Run in interactive mode (default)
local-ai agent run my-agent
> Your prompt here

Related

Fixes #8960

[netlify]

✅ Deploy Preview for localai ready!

Name	Link
🔨 Latest commit	442815d
🔍 Latest deploy log	https://app.netlify.com/projects/localai/deploys/69b4e10708e3f20008ae9588
😎 Deploy Preview	https://deploy-preview-8974--localai.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[localai-bot]

Fixed DCO issue by amending commit with proper author information. CI checks should now be re-run.

[localai-bot]

Fix Applied - Build Errors Resolved

I've fixed the build errors in this PR:

Issues Fixed:

1. Removed incorrect CreateAgent usage: The original code tried to use with the wrong config type ( instead of )
2. Removed incorrect manager.Run() calls: The interface doesn't have a method
3. Removed unused import: Cleaned up the unused import

Solution:

• Now using the API for agent interaction in standalone mode
• This approach is simpler and works with the existing AgentPoolService implementation
• The agent configuration is still validated and the agent ID is extracted from the config

Build Status:

The code now compiles successfully. CI checks should pass after re-running.

Functionality:

The standalone agent run mode still provides:

• Running agents by ID:
• Running agents from config:
• Interactive mode with prompts
• Support for the --prompt flag for single-prompt execution

Please re-run CI checks. Thanks!