fix: enable reasoning visibility for DeepSeek V3.1 and GLM-4.5 models

[roomote]

This PR fixes the reasoning/thinking section visibility issue for DeepSeek V3.1 and GLM-4.5 models when used through OpenAI Compatible provider.

Problem

• DeepSeek V3.1 and GLM-4.5 models' reasoning sections were not appearing when reasoning effort was enabled
• Both models support thinking/reasoning features but lacked the necessary configuration flags

Solution

• Added supportsReasoningEffort: true flag to DeepSeek chat and reasoner models
• Added supportsReasoningEffort: true flag to all GLM-4.5 model variants (standard and Air versions)
• Applied to both international and mainland configurations for GLM models

Testing

• All existing tests pass
• Type checking passes
• Linting passes

Impact

This change enables users to see the reasoning/thinking process for:

• DeepSeek V3.1 (deepseek-chat)
• DeepSeek R1 (deepseek-reasoner)
• GLM-4.5 (all variants)
• GLM-4.5-Air (all variants)

Fixes #7370

cc @AykhanUV - Yes, this fix addresses both DeepSeek V3.1 and GLM-4.5 models as requested.

---

Important

Adds supportsReasoningEffort: true to enable reasoning visibility for DeepSeek V3.1, DeepSeek R1, GLM-4.5, and GLM-4.5-Air models.

• Behavior:
  • Adds supportsReasoningEffort: true to deepSeekModels in deepseek.ts for DeepSeek V3.1 and DeepSeek R1 models.
  • Adds supportsReasoningEffort: true to internationalZAiModels and mainlandZAiModels in zai.ts for GLM-4.5 and GLM-4.5-Air models.
• Impact:
  • Enables reasoning/thinking visibility for DeepSeek V3.1, DeepSeek R1, GLM-4.5, and GLM-4.5-Air models.
  • Applies to both international and mainland configurations for GLM models.

^{This description was created by for 774afd5. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Reviewing my own code because apparently I trust no one, not even myself.

[roomote]

The implementation looks correct and follows the existing pattern. However, would it be helpful to add test coverage for the supportsReasoningEffort flag? The existing tests in src/api/providers/__tests__/deepseek.spec.ts could verify this capability is properly set.

[roomote]

Minor suggestion: For consistency, consider standardizing the comment wording across all models. Currently using "DeepSeek V3.1 supports thinking/reasoning" here and "DeepSeek-R1 supports thinking/reasoning" below. Perhaps just "supports reasoning" would be cleaner?

[roomote]

Good to see both international and mainland configurations updated consistently. Similar to the DeepSeek models, consider adding test assertions in src/api/providers/__tests__/zai.spec.ts to verify the supportsReasoningEffort flag is properly set for these models.

[daniel-lxs]

Issue needs scoping