fix(proxy): wire LLM latency metrics into the streaming response path (#63)

`NewLLMMetricsCapture`/`ResponseWriterWrapper` had **zero call sites**, so `llm_ttft_seconds`, `llm_itl_seconds` and `llm_tokens_total` were exported as HELP/TYPE stubs with no samples (verified against the live `/metrics`).

These are the only per-request LLM latency series covering the Ollama models (`ornith:35b`, `qwen2.5:3b-instruct`), which expose no `/metrics` of their own — so those models had zero latency visibility.

## Change

Wire the wrapper around the response writer in `ServeHTTP` for model-routed requests, and delete the unused `io.WriteCloser` duplicate that made the gap hard to spot (two near-identical implementations, neither constructed).

## Correctness details

- **Only SSE responses are measured.** A non-streaming completion is written in one `Write`, which would otherwise be recorded as a TTFT sample equal to the whole request latency, with no ITL samples — plausible-looking but meaningless numbers.
- **`requestStart` is captured at the top of `ServeHTTP`** so TTFT measures from request acceptance, not from when the upstream response is copied.
- **Token count skips the `data: [DONE]` terminator.**
- **`Unwrap()`** keeps `http.ResponseController` (Flush/Hijack/deadlines) working through the wrapper, so SSE unbuffering (#33) is unaffected.

## Tests

| Test | Guards |
|---|---|
| `TestResponseWriterWrapperRecordsSSEMetrics` | TTFT=1, ITL>0, tokens=2 (terminator excluded) |
| `TestResponseWriterWrapperIgnoresNonStreaming` | JSON response records nothing |
| `TestResponseWriterWrapperPreservesStreaming` | bytes + Flush pass through |
| `TestHandlerRecordsLLMMetricsForStreaming` | end-to-end through `ServeHTTP` |

The last one fails if the wrapper is ever wired nowhere again — verified by disabling the wiring locally and watching it fail.

## Verification needed post-merge

Confirm samples appear:
```
curl -s https://api.riotpiao.com/metrics | grep -E "llm_(ttft|itl)_seconds_count|llm_tokens_total"
```

internal/proxy/proxy.go

@@ -314,6 +314,10 @@ func writeProblemDetail(w http.ResponseWriter, status int, problemType, title, d
 
 // ServeHTTP implements http.Handler.
 func (h *Handler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
+	// Captured before any work so TTFT measures from the moment the gateway
+	// accepted the request, not from when the upstream response is copied.
+	requestStart := time.Now()
+
 	if h.handleAuthEndpoint(w, r) {
 		return
 	}
@@ -358,7 +362,14 @@ func (h *Handler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
 	// For streaming responses (SSE, chunked), disable buffering. Issue #33.
 	w.Header().Set("X-Accel-Buffering", "no")
 
-	proxy.ServeHTTP(w, r)
+	// Capture per-request LLM latency metrics (TTFT/ITL/tokens) while the
+	// response is copied from the upstream. Inert for non-SSE responses.
+	var respWriter http.ResponseWriter = w
+	if route.ModelName != "" {
+		respWriter = NewResponseWriterWrapper(w, route.ModelName, observability.Global(), requestStart)
+	}
+
+	proxy.ServeHTTP(respWriter, r)
 }