#1943: Add jan.ai provider documentation

added jan.ai provider integration with auto-discovery - Add jan.ai provider implementation with automatic model discovery - Update documentation to include jan.ai setup and configuration - Extend FAQ to mention jan.ai as supported self-hosted option - Update tsconfig to include Node.js types for type safety <!-- greptile_comment --> <h2>Greptile Overview</h2> <h3>Greptile Summary</h3> This PR adds a new `jan` model provider with local auto-discovery (fetching `/v1/models` from a default local jan.ai base URL), plus documentation for setup/configuration and a FAQ update. It also tweaks CI/Vitest worker/timeouts and the parallel test runner to improve stability. The new provider wiring follows the existing implicit-provider pattern (only enabled when auth is present), but there is a correctness issue in the env-key resolution path for `jan` that can cause the provider `apiKey` to be set incorrectly. Additionally, the test runner’s Node memory flag is currently passed in a way that likely has no effect. <h3>Confidence Score: 3/5</h3> - This PR is moderately safe to merge, but has a correctness issue in jan provider auth resolution that should be fixed first. - Most changes are documentation and CI/test tuning, but the jan provider wiring likely assigns an identifier string instead of a resolved credential when using env-based configuration, which can break provider availability/discovery in normal usage. The test runner memory flag change also appears ineffective as implemented. - src/agents/models-config.providers.ts, scripts/test-parallel.mjs <!-- greptile_other_comments_section --> <!-- /greptile_comment -->