feat: Add CLI flags for mini: step-limit, temperature, timeout, and max-tokens

[Chesars]

Summary

Adds four new CLI flags to mini command for better control over agent execution and model behavior:

• --step-limit: Maximum number of steps the agent can take (set to 0 for no limit)
• --temperature: Model temperature control (0.0-1.0, lower is more deterministic)
• --timeout: Command execution timeout in seconds
• --max-tokens: Maximum tokens in model response

Details

Add 4 new CLI flags with proper type annotations in src/minisweagent/run/mini.py
Organize config overrides into three clear sections: Agent, Environment, and Model
All flags properly override config file values when specified

Testing

✅ All 28 CLI integration tests passing

Usage Examples

Limit agent to 50 steps with lower temperature for more deterministic output
mini --step-limit 50 --temperature 0.3

Set command timeout and limit model response tokens
mini --timeout 120 --max-tokens 8000

Disable step limit completely
mini --step-limit 0

Complex example: Run agent with full control over execution parameters

mini \
  --model claude-sonnet-4-5-20250929 \
  --step-limit 100 \
  --temperature 0.2 \
  --timeout 300 \
  --max-tokens 8000 \
  --cost-limit 5.0 \
  --yolo \
  --task "Refactor the authentication module to use OAuth2"

Disable limits for unrestricted execution
mini --step-limit 0 --cost-limit 0

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/minisweagent/run/mini.py	88.40% <100.00%> (+3.15%)	⬆️
tests/run/test_cli_integration.py	100.00% <100.00%> (ø)
tests/run/test_local.py	100.00% <ø> (ø)

... and 2 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.