MCTS Reasoning Agent

[BabyCNM]

Why are these changes needed?

Related issue number

TODOs

• Add test case
• Add image for the blog
• Improve the writing for the blog and the notebook.
• Add the LATS approach.
• Add reflection to the nodes.
• remove the redundant methods from mcts and lats.
• Add test cases to LATS
• Benchmark the 3 approaches.
• Separate llm_config for the evaluator
• Evaluation normalization and different scales (instead of 1 - 5, how about 1 - 100?)

Modularization

• MCTS
• LATS
• P1 Agent to save cost O(depth) with merging thinker and grader
• Forest of Thoughts

Engineering

• [] Async Client call
• [] Swarm Agent implementation

Checks

• I've included any doc changes needed for https://ag2ai.github.io/ag2/. See https://ag2ai.github.io/ag2/docs/Contribute#documentation to build and test documentation locally.
• I've added tests (if relevant) corresponding to the changes introduced in this PR.
• I've made sure all auto checks have passed.

Acknowledgement

We appreciate the feedbacks and suggestions from Yaron Been and namastex888.

[Hk669]

@sonichi do we need ReasoningAgent Test in the workflows??

[marklysze]

Having run this through for the demo and read through the blog and notebook, I think it's good to go!

Nice work team, it's great to have the extra capability for the ReasoningAgent :).

[andyz245]

Commenting on the LATS integration - it looks good to me!