Survey automation

[dimits-ts]

Fixes #413

Added basic support for LLM participants to answer surveys.

Features:

• Prompt structure for each question type
• Basic unit tests that ensure all relevant information stays included in the prompt (useful to avoid bugs during refactoring)
• Connected implementation with @vivtsai 's experimental agent testing function

The implementation currently does not validate the LLM output nor does it coerce it to a type readable by the backend. Let me know if this is within the scope of this PR, and I'll add it.

• Tests pass
• Appropriate changes to documentation are included in the PR

Tested with llama3.2-1B. If it runs correctly with such a small model, I reckon it's safe to assume much larger LLMs won't have any issues.

[vivtsai]

The implementation currently does not validate the LLM output nor does it coerce it to a type readable by the backend. Let me know if this is within the scope of this PR, and I'll add it.

This is within scope, so feel free to add here or in a follow-up PR! :) @dimits-ts