feat: add MiniMax as LLM provider with M2.7 model support

[octo-patch]

Summary

• Add MiniMax as a first-class LLM provider via OpenAI-compatible eino-ext integration
• Support MiniMax-M2.5, M2.5-highspeed, M2.7, and M2.7-highspeed models
• M2.7 is the latest model with advanced reasoning capabilities and 16K output tokens
• Temperature clamping (MiniMax requires temperature > 0)
• Response format forced to text (MiniMax does not support JSON mode)
• Full model builder with unit and integration tests

Changed Files

• backend/bizpkg/llm/modelbuilder/minimax.go - MiniMax model builder
• backend/bizpkg/llm/modelbuilder/minimax_test.go - Unit + integration tests
• backend/bizpkg/llm/modelbuilder/model_builder.go - Registration
• backend/conf/model/model_meta.json - Model metadata with M2.5/M2.7 entries
• backend/conf/model/template/model_template_minimax.yaml - Template config (M2.7 default)
• helm/charts/opencoze/files/conf/model/model_meta.json - Helm chart model metadata
• Multiple auto-generated IDL files with ModelClass_MiniMax enum

Test Plan

• Unit tests for temperature clamping
• Unit tests for model builder config
• Unit tests for model builder registration
• Integration tests for M2.5, M2.5-highspeed, M2.7, M2.7-highspeed models
• JSON and YAML file validation

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

✅ octo-patch
❌ PR Bot

---

PR Bot seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}