refactor: cortex talks to neurons instead of mistral.rs directly

Replace NodeConfig (static vram_mb, pinned) with NeuronEndpoint.
Hardware discovery and model pinning now come from neuron API and
models.toml catalogue respectively.

- config.rs: nodes -> neurons, add models_config path
- catalogue.rs: ModelProfile with pinned_on, ModelCatalogue
- poller.rs: poll neuron GET /models (ModelInfo format)
- router.rs: resolve inference endpoint via neuron GET /models/{id}/endpoint
- evictor.rs: call neuron POST /models/unload
- node.rs: remove vram_mb, pinned fields (come from discovery/catalogue)
- All 22 gateway tests updated to mock neuron API
- Remove MistralModelsResponse, ModelLifecycleRequest (no longer needed)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Cargo.lock

@@ -401,6 +401,7 @@ dependencies = [
  "tower",
  "tower-http",
  "tracing",
+ "urlencoding",
 ]
 
 [[package]]
@@ -2237,6 +2238,12 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "urlencoding"
+version = "2.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "daf8dba3b7eb870caf1ddeed7bc9d2a049f3cfdfae7cb521b087cc33ae4c49da"
+
 [[package]]
 name = "utf8_iter"
 version = "1.0.4"