feat: add Claude Opus 4.8 model

[tasoo-oos]

PR Checklist

• Required Checks
  • Have you added type definitions?
  • Have you tested your changes?
  • Have you checked that it won't break any existing features?
• If your PR uses models¹, check the following:
  • Have you checked if it works normally in all models?
  • Have you checked if it works normally in all web, local, and node-hosted versions? If it doesn't, have you blocked it in those versions?
• If your PR is highly AI generated², check the following:
  • Have you understood what the code does?
  • Have you cleaned up any unnecessary or redundant code?
  • Is it not a huge change?
    • We currently do not accept highly AI generated PRs that are large changes.

Summary

Add Claude Opus 4.8 to the Anthropic model list and add the current AWS Bedrock Claude Opus 4.7 model ID.

Related Issues

None.

Changes

• Add claude-opus-4-8 as a recommended Anthropic model.
• Configure Opus 4.8 with adaptive thinking and xhigh effort support.
• Omit unsupported sampling parameters for Opus 4.8.
• Expose the claudeXHighEffort custom flag in model settings.

Impact

Users can select Claude Opus 4.8 through the RisuAI Official Anthropic API implementation. Existing models are unchanged.

Additional notes

Manually tested with streaming on / off and various thinking settings.

Footnotes

1. Modifies the behavior of prompting, requesting, or handling responses from AI models. ↩

2. Over 80% of the code is AI generated. ↩

[cubicj]

One Bedrock-specific issue: this adds claudeAdaptiveThinking to AWS Claude 4.7 Opus, so the existing adaptive thinking UI can expose effort: "max" for it. AWS documents max as Claude Opus 4.6-only, while other models should use low, medium, or high.

https://docs.aws.amazon.com/bedrock/latest/userguide/claude-messages-adaptive-thinking.html

Could you narrow the exposed effort options for Bedrock Opus 4.7?

[tasoo-oos]

I removed AWS support for Claude Opus 4.7 from this PR.

At first, I thought it won't be so troublesome since there's already an API implementation for AWS and new model's behavior seems to be the same as the last one (max adaptive thinking behavior change part was not mentioned in model docs), so It would be okay to just slip it into the PR.

However, if model behavior has changed and I am unable to test it, I think it's better not to add it.

Besides, who would possibly try to contact AWS sales to use Opus 4.7 for RolePlay when they could just go to Anthropic or Openrouter?

But at the same time, I should have searched for the docs more rigorously, or be more conservative with what I cannot test by myself. Sorry about that.