feat: wire the tier through the transport and keep each tier's warning distinct

The transport now hands back the tier it actually sent, and that tier
rides TransportResult into LLMResponse. It is not the requested one:
under EFFORT_FALLBACK=nearest a medium request goes out as low, and
telemetry grouping by the requested tier would file the row under a tier
that never left the process.

Reconciliation judges the same tier instead of the old enable_thinking
bool, and the warning throttle keys on it. Keyed on the bool, every tier
of one model shared a single key, so the second contradiction was
silenced for the lifetime of the transport. The predicate is an identity
check against Effort.NONE on purpose -- the member's value is the
non-empty string "none", so any truthiness test would send every strength
tier down the "asked to disable" branch and invert the alarm.

src/polygateway/types.py

@@ -326,6 +326,16 @@ class LLMResponse:
     实测开启档 completion 53 vs 关闭档 3),该档即为 `UNKNOWN`。
     要判"确实没推理"只认 `ABSENT`(上游明确上报 0)。"""
 
+    applied_effort: Effort | None = None
+    """本次调用**真正发出去**的推理档位(issue #20);`None` = 调用方未表态。
+
+    与 `ChatRequest.reasoning_effort`(请求档)可能分叉: 源上配了
+    `EFFORT_FALLBACK=nearest` 时,请求 `medium` 而模型只有 low/high/max,实际发
+    出的是 `low`。遥测按本字段分组,记请求档会把整行挂在一个从未发出过的档下。
+
+    `None` 不是"没推理": 库不表态时也不推定模型自己的默认档——"没看见"不许说成
+    "发生了"(同 `thinking_observation` 的 `UNKNOWN` 一脉)。"""
+
 
 @dataclass(frozen=True)
 class ChatRequest:
@@ -438,6 +448,13 @@ class TransportResult:
     默认 `UNKNOWN` 而非 `ABSENT`: 不做裁定的 transport(OCR/embedding 等)沉默
     时,不该替上游做出"没推理"这个它从未做过的声明。"""
 
+    applied_effort: Effort | None = None
+    """本次调用真正发出去的推理档位(issue #20),由做注入的 transport 填。
+
+    只有做了注入的那一层知道它: `nearest` 映射后请求档与实际档分叉(请求
+    `medium` → 实发 `low`),中间件事后再算一遍必然算成请求档。默认 `None` 是
+    "未表态/不注入推理参数"(OCR、embedding 等 transport 沉默即此)。"""
+
 
 @dataclass(frozen=True)
 class SourceConfig: