feat(tools): add MiniMax image generation backend

[octo-patch]

Reason: Add direct MiniMax text-to-image generation with regional endpoints and complete response handling.

Summary

• Add explicit MiniMax image provider and region selection with global and China endpoints.
• Send supported text-to-image parameters and persist URL or base64 results in the workspace.
• Reuse configured MiniMax API credentials while preserving the existing image path.

Type

• Fix
• Feature
• Docs
• CI / tooling
• Refactor
• Other

Verification

• Relevant tests pass locally: uv run pytest tests/test_media_gen_tool.py tests/test_config_provider_passthrough.py -x -q (27 passed)
• Relevant lint / type checks pass locally: uv run ruff check raven/agent/tools/media_gen.py raven/config/schema.py tests/test_media_gen_tool.py
• User-facing docs or screenshots are updated when needed

Risk

• Security impact considered
• Backward compatibility considered
• Rollback path is clear for risky changes

The API credential is sent only to the configured generation endpoint and is not forwarded when downloading generated image URLs. Existing image configurations continue to use their current provider. Rollback is a single commit revert.

Related Issues

N/A