feat: add skills field to GenerationConfig and bump version to 0.5.10

[devin-ai-integration]

Summary

Adds an optional skills: Optional[List[AgentSkill]] field to GenerationConfig, matching the field already present in AgentExecutionOrCreationConfig. This allows skills to be passed in prediction requests (image, document, video, audio), not just agent execution/creation requests.

Also bumps the SDK version from 0.5.9 → 0.5.10.

Review & Testing Checklist for Human

• Verify the backend API accepts skills in the generation config payload for prediction endpoints (/image/generate, /document/generate, etc.). The SDK will now include "skills": null (or the list) in the serialized config sent to the API — confirm this doesn't break existing endpoints that don't expect this field.
• Check model_dump() serialization: GenerationConfig.model_dump() doesn't have special handling for skills (unlike response_model→json_schema conversion). Confirm that Pydantic's default serialization of List[AgentSkill] produces the correct JSON structure the API expects.
• Confirm version bump is appropriate: 0.5.9 → 0.5.10 for a minor feature addition.

Notes

• The skills field defaults to None, so this is backward compatible for all existing callers.
• No new tests were added. Consider adding a unit test for GenerationConfig(skills=[AgentSkill(...)]).model_dump().

Link to Devin session: https://app.devin.ai/sessions/f0720ba9140f4c089f1deaa4894bb1f9

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring