feat(brain): add Cheaper Inference LLM provider

[aiapienthusiast]

Summary

• This PR adds Cheaper Inference as an opt-in LLM provider for standalone bughunter / brain.py.
• It uses the same pattern as feat(brain): add Requesty LLM provider #159 (Requesty).

Cheaper Inference is an OpenAI-compatible LLM gateway.
One API key gives access to models from several labs.

Usage:

export CHEAPER_INFERENCE_API_KEY=ci_live_...
bughunter setup --provider cheaperinference   # or option 12, or BRAIN_PROVIDER=cheaperinference

Type

• Bug fix
• New feature / scanner module
• Methodology improvement
• Documentation
• False positive reduction

Test Plan

• pytest tests/ passes locally
• No hardcoded targets, API keys, or real domain names
• Scanner additions use [CONFIRMED] / [POSSIBLE] / [INFORMATIONAL] confidence states
• New functionality has at least one regression test

Details

Files:

• bughunter/brain.py: provider init, chat dispatch, default model gpt-5.4-mini, curated model list, docstring.
• bughunter/engine.py: setup option 12, providers table, --provider help and choices, saved key loading.
• config.example.json, README.md, CHANGELOG.md: docs.
• tests/test_brain_auto_detect.py: 3 new tests (key jumps to front, init sets base URL and auth header, not in PROVIDER_PRIORITY).
• tests/test_brain_litellm.py: clear the new env var.

Behavior:

• Base URL: https://api.cheaperinference.com/v1.
• Key env var: CHEAPER_INFERENCE_API_KEY.
• The provider is not added to PROVIDER_PRIORITY, like Requesty.

Tests:

• pytest -q: 977 passed.
• The tests use mocks.
• Live check: gpt-5.4-mini returned a reply through LLMClient with BRAIN_PROVIDER=cheaperinference.
• Live check: all curated model ids are in the live /v1/models list.