[nicho810]

What this adds

A cheapaimodel preset in the LLM spec list, so a model behind cheapaimodel can be picked as a scan target in the UI and through GET /v1/llm-specs, the same way the Runcrate preset added in #361 works. cheapaimodel is an OpenAI-compatible chat completions gateway at https://api.cheapaimodel.ai/v1 (models include DeepSeek, GLM, Kimi and Grok).

The preset sends <<PROMPT>> as the user message to deepseek-v4-1-flash, with Authorization: Bearer $CHEAPAIMODEL_API_KEY.

Files (same shape as #361):

• agentic_security/routes/_specs.py: spec appended to _SPECS and a matching cheapaimodel entry appended to LLM_CONFIGS (same position, so the zip that builds LLM_SPECS stays aligned). It uses the existing fallback icon, /icons/myshell.png.
• agentic_security/static/base.js: the same spec and config appended to the UI lists.
• tests/unit/test_cheapaimodel_spec.py: the preset parses to a POST to https://api.cheapaimodel.ai/v1/chat/completions with the right model and text modality, and probe() sends the prompt as a valid JSON user message (the HTTP call is stubbed).

Checks run

• pytest tests/unit (Python 3.15, poetry install): 355 passed, 1 skipped.
• black --check and flake8 on the changed Python files: clean.
• Unauthenticated check: GET https://api.cheapaimodel.ai/v1/models without a key returns 401. I did not make a live model call from this branch.

Disclosure: I work on cheapaimodel.

[msoedov]

@nicho810 thx for the patch