🤖 feat: Add Z.AI GLM Context Window & Pricing

[danny-avila]

Summary

I added comprehensive support for Zhipu AI's GLM model family, including context window configurations and pricing multipliers for all GLM variants.

• Added context window limits for all GLM models (glm-4, glm-4-32b, glm-4.5, glm-4.5-air, glm-4.5v, glm-4.6) ranging from 66K to 200K tokens
• Implemented pricing multipliers for prompt and completion tokens across all GLM model variants
• Enhanced model name matching to handle provider prefixes (z-ai/, zai/, zai-org/) and suffixes (-fp8, -FP8) with case-insensitive matching
• Refactored findMatchingPattern function to use case-insensitive comparison for improved pattern matching reliability
• Added comprehensive test coverage for GLM models including pattern matching, token limits, and pricing calculations across various naming conventions
• Updated @librechat/agents to v2.4.83 to address a reasoning edge case encountered with GLM models

Change Type

• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)

Testing

Tested GLM model support including context window limits, pricing calculations, and model name matching with various provider prefixes and suffixes. Verified case-insensitive matching works correctly and that all model variants return expected token limits and pricing multipliers.

Test Configuration

• Node.js environment with Jest test runner
• Test cases cover exact model names, provider-prefixed names, case variations, and suffix variations
• Validated against tokenValues and maxTokensMap configurations

Checklist

• My code adheres to this project's style guidelines
• I have performed a self-review of my own code
• I have commented in any complex areas of my code
• My changes do not introduce new warnings
• I have written tests demonstrating that my changes are effective or that my feature works
• Local unit tests pass with my changes
• Any changes dependent on mine have been merged and published in downstream modules