Add support for reasoning_effort parameter for reasoning models in AzureOpenAIConfig

[MRizwan14]

🚀 The feature

A feature for detecting reasoning models such as o1, o3 and gpt-5 was recently added within the file https://github.com/mem0ai/mem0/blob/main/mem0/llms/base.py through the _is_reasoning_model() method.

However, the newly supported OpenAI SDK parameter reasoning_effort was not implemented in Mem0’s configuration classes. As a result, when "reasoning_effort": "low" is included in the llm.config section (for example, when initializing via Memory.from_config), the following error is raised: TypeError: AzureOpenAIConfig.__init__() got an unexpected keyword argument 'reasoning_effort'

Motivation, pitch

This feature is requested to enable testing and comparison of different reasoning effort levels ("low", "medium", "high") supported by the latest OpenAI SDK. Adding this parameter would make it possible to evaluate performance and latency trade-offs across reasoning models directly within Mem0.

[parshvadaftari]

Hey @MRizwan14 Thank you for pointing this out, definitely this will be a great new addition for the mem0.

[ron-42]

Hi @parshvadaftari I want to work on this issue, could you please assign it to me?

[Anand0295]

Hi, I'm a beginner interested in Python, AI/ML, and would love to work on this. Please assign it to me!

[agam1092005]

I would like to work on this !

[shubham2345]

Hello @parshvadaftari , could you assign this issue to me.