Implement language detection and response handling in the agent

[n1colasf]

Description

Fixes inconsistent response language by adding language detection before generation and enforcing a session language policy based on the first user message. Also adds safe fallbacks for detection and translation failures, updates prompt guidance, and documents setup and behavior.

Type of Change

• Bug fix (non-breaking change that fixes an issue)

Related Issues

Fixes #1

Changes Made

• Added a new pretool named detect_user_language to detect user language from chat history and inject runtime language context before response generation.
• Implemented pinned session language behavior using the first user message, with fallback to last user message and then English when detection is unreliable.
• Added dynamic translation for the no-information message with safe fallback to the original message if translation fails.
• Registered the language detection tool in the agent pretools pipeline.
• Updated system prompt rules to explicitly follow runtime language context and keep language stable unless user requests a switch.
• Added dependency declarations for pycld2 and translate.
• Updated README with dependency installation and language behavior documentation.

Testing Done

• Tested manually

Checklist

• My code follows the project's style guidelines
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings