feat: let one call ask for a different tier than its source defaults to

The three-layer priority (call > source > enable_thinking sugar > silence)
now lives in one pure function, thinking.effective_effort(). The assembly
guard and the request hot path used to each carry their own inline copy of
the sugar conversion; two copies of the same judgement drift into the worst
shape there is - passes at assembly, raises at runtime.

The guard now also honours effort_fallback, so a source that opted into
nearest is no longer sentenced at assembly for a tier it could have mapped.

src/polygateway/client.py

@@ -34,7 +34,7 @@ from polygateway.sources import (
     RoundRobinSelector,
     SourceCooldownMemo,
 )
-from polygateway.thinking import get_capability, resolve_thinking
+from polygateway.thinking import effective_effort, get_capability, resolve_thinking
 from polygateway.transports.openai_compat import OpenAICompatTransport
 from polygateway.types import (
     ChatRequest,
@@ -81,16 +81,20 @@ def _guard_thinking(
     就带着指路信息炸掉。`get_provider` 现在就是同一形态的双点调用。
     """
     for source, profile in zip(sources, profiles, strict=True):
-        # `enable_thinking` 的档位语法糖(True → auto,False → none,None 不表态);
-        # 两个调用点各自就地转换是过渡形态,T5 起由 thinking.effective_effort()
-        # 统一收口并接上源级/请求级档位(设计 §4.2)
-        enabled = source.enable_thinking
-        effort = None if enabled is None else (Effort.AUTO if enabled else Effort.NONE)
         resolve_thinking(
             profile,
             get_capability(source.model, table=capabilities),
-            effort,
+            # 装配期看不见请求级档位(它逐次调用才产生),故只解源级两层;请求级
+            # 只能在运行期由 transport 校验(设计 §10 的装配期/运行期分工)
+            effective_effort(
+                request_effort=None,
+                source_effort=source.reasoning_effort,
+                enable_thinking=source.enable_thinking,
+            ),
             model=source.model,
+            # 与 transport 用同一个 fallback,否则配了 nearest 的源会在装配期就被
+            # 判死,而它在运行期本来是能映射到最近档跑起来的
+            fallback=source.effort_fallback,
         )
 
 
@@ -295,6 +299,7 @@ class GatewayClient:
         structured: type[BaseModel] | Literal["json"] | None = None,
         stream: bool = True,
         overlay: Mapping[str, Any] | None = None,
+        reasoning_effort: Effort | None = None,
         tenant_id: str | None = None,
         meta: Mapping[str, Any] | None = None,
     ) -> LLMResponse:
@@ -304,6 +309,10 @@ class GatewayClient:
         高于源级 `extra_body`、低于结构化输出的注入。带默认值的 keyword-only
         参数不影响既有调用点(issue #4)。
 
+        `reasoning_effort` 是本次调用的推理档位,优先级高于源级 `REASONING_EFFORT`
+        与 `ENABLE_THINKING`(设计 §4.2)。`None` 是**不表态**(随源级配置),与
+        `Effort.NONE`("要求不推理")严格区分。
+
         `tenant_id` 与 `meta` 是调用方自定义维度,只进遥测、**不进缓存 key**
         (租户隔离由 `cache_namespace` 负责,ARCH §7.5);前者享有真实列待遇
         (可挂 RLS、可进复合索引),后者是任意 KV 容器(issue #11)。
@@ -333,6 +342,7 @@ class GatewayClient:
             stream=stream,
             overlay=sampling,
             sampling=sampling,
+            reasoning_effort=reasoning_effort,
             tenant_id=dimension_tenant_id,
             meta=dimensions,
         )