feat: cortex-cpp: endpoint to returns all models

[vansangpfiev]

cortex-cpp returns a list of running models, so we don't have to send every single request over the full models list (let's say it's 1000 models)