[BUG] Camel training example code is not passing important parameters.

[HwVanICI]

Checklist

• The error occurs when using our provided Docker image.
• I can consistently reproduce the bug across multiple trials or random seeds.
• [] If the error causes experiment abortion, I've verified that this error is the root
  cause, not a secondary error caused by peer workers.

Detailed Information

Describe the bug

In the Camel example training code, the run_agent() function of CamelMathAgent class does not pass two important parameters to CamelAgent class as follows.

• max_tokens parameter for model configuration is not passed to ChatAgent class properly. It is passed via "token_limit" parameter which was not accepted by CamelAgent class. As a result, the CamelAgent is using default max_tokens value of 512 and it is too short and finally affecting wrong reward values during training.
• step_timeout parameter is not passed to CamelAgent, and it is causing the CamelAgent use default step_timeout of 180. And it is too short for some samples and disrupting the training.

Expected behavior

We want to:

1. Remove the "token_limit" parameter which is not accepted by CamelAgent.
2. Create a new "model_config_dict" dictionary including "max_tokens" elememnt, and pass it to the model parameter when creating ChatAgent instance.
3. Pass the "step_timeout" parameter when creating a ChatAgent instance.
4. Added additional rollout configuration in the config.yaml file to set "request_timeout" which can be used to set the "step_timeout".