refactor: make capability a tier list, since "can it be off" is one entry

The boolean could say a model reasons or does not. It could not say what
GLM-5.3 and Gemini 3 Pro actually do: refuse to stop reasoning while
still letting you ask for less. So capability becomes the list of tiers a
model serves, and `none`'s presence in it is what "can_disable" now reads.

Effort carries `auto` alongside the strength tiers. Nine of the models on
our gateway are pure switches with no tier to name, and without `auto`
they would have to borrow a strength tier to mean "on" — which is the
exact bug this work exists to remove.

Tiers land as documented guesses from four registries that agree; every
entry says so in its evidence, and task 10 replaces them with measurements.

src/polygateway/__init__.py

@@ -32,6 +32,8 @@ from polygateway.thinking import (
     resolve_thinking,
 )
 from polygateway.types import (
+    EFFORT_ORDER,
+    Effort,
     EmbeddingResponse,
     LLMResponse,
     OcrLayoutElement,
@@ -46,6 +48,8 @@ __version__ = "1.3.2"
 
 __all__ = [
     "DEFAULT_PROFILES",
+    "EFFORT_ORDER",
+    "Effort",
     "AllSourcesExhausted",
     "CircuitOpenError",
     "EmbeddingClient",