feat: upgrade MiniMax default model to M2.7

[octo-patch]

Summary

• Add MiniMax-M2.7 and MiniMax-M2.7-highspeed to the model selection list
• Set MiniMax-M2.7 as the new default model
• Retain all previous models (M2.5, M2.5-highspeed, M2.1) as available alternatives
• Update related unit tests and documentation

Changes

• agent/model_metadata.py: Add M2.7 context length entries
• agent/auxiliary_client.py: Update default auxiliary model to M2.7-highspeed
• agent/insights.py: Add M2.7 pricing entries
• hermes_cli/models.py: Add M2.7 to provider model lists, update OpenRouter entry
• hermes_cli/setup.py: Add M2.7 to setup wizard model selection
• tools/rl_training_tool.py: Update test inference model to M2.7
• tests/test_fallback_model.py: Update fallback test to use M2.7
• docs: Update configuration examples and RL training docs

Why

MiniMax-M2.7 is the latest flagship model with enhanced reasoning and coding capabilities.

Testing

• All changes follow existing patterns exactly (model list additions + default updates)
• Unit tests updated to reflect M2.7 as the recommended model

[teknium1]

Merged via PR #1900. Your MiniMax M2.7 changes were applied manually onto current main (the pricing module was refactored since your branch, so cherry-pick had conflicts). Authorship preserved on the commit. We also added several new OpenRouter models in the same PR. Thanks!