feat(cag): CAG integration — memory adapter, warmup semaphore, cache-aware routing (#47)

## Summary

Three issues in one PR. Implements Cache-Augmented Generation (CAG) support in the api-gw.

---

### #44 — Memory ServiceAdapter: standing-context resource

- Code-registers `memory` adapter with `standing-context` resource (`POST /memory/standing-context`)
- Fixes dispatcher bug: method-level `Auth.Capability` override now respected (was always using verb-derived capability)
- Auth: `memory:read` required via method-level override; `memory:write` for ingest
- `MEMORY_UPSTREAM_URL` env configures upstream

### #45 — Continuous batching: warmup deprioritisation + per-backend semaphore

- In-memory bounded queue per backend: `VLLM_MAX_QUEUE_DEPTH` slots (default 20)
- Concurrency limiter: `VLLM_MAX_NUM_SEQS` slots (default 4)
- `X-Poimen-Warmup: true` → sleep `WARMUP_DELAY_MS` (default 500ms) before queue entry, strip header, add `X-Priority: low`
- Queue full → 503 immediately, no goroutine pile-up
- Metrics: `warmup_requests_delayed_total`, `queue_full_requests_total` at `/metrics`

### #46 — Cache-aware routing: rendezvous hash + prefix cache hit logging

- `addresses: [...]` YAML field for multi-backend models (backward-compat with `address`)
- `X-Poimen-Cache-Key` header → rendezvous-hash backend selection (same key = same GPU)
- No cache key → round-robin
- Failed backend excluded 30s; next-highest-score fallback
- `vllm-num-cached-tokens` response header captured → `vllm_prefix_cache_hit_tokens_total` counter + structured log

---

## Verification

```
go build ./...    ✅
go vet ./...      ✅
go test -race ./... ✅ (all packages, 0 races)
```

## Reviewer verdicts

- reviewer-sre: SIGN-OFF
- reviewer-quality: SIGN-OFF

## Non-blocking findings (🟡🔵 only)

- `getOrCreateTransport` not goroutine-safe by design; pre-population invariant documented
- `time.Sleep(warmupDelay)` holds goroutine; context-cancel alternative left as TODO
- OTel trace_id not propagated into CAG log events (existing gap)

Closes #44 Closes #45 Closes #46

---------

Co-authored-by: Admin Bot <[email protected]>
Reviewed-on: #47

internal/config/config.go

@@ -33,7 +33,11 @@ type ModelUpstream struct {
 	// Name is the model name clients send (e.g., "reasoning", "ornith:35b").
 	Name string
 	// Address is the upstream server address (host:port).
+	// When Addresses is set, this holds Addresses[0] for backward compatibility.
 	Address string
+	// Addresses holds all backend addresses for multi-backend cache-aware routing.
+	// Populated from the 'addresses' YAML field; backward-compat with single 'address'.
+	Addresses []string
 	// Path is the upstream path for this model (e.g., "/v1/chat/completions").
 	Path string
 	// UpstreamModel is the model name to send to the upstream server.
@@ -116,6 +120,18 @@ func (c *Config) ModelNames() []string {
 	return names
 }
 
+// AllAddresses returns all configured backend addresses for the model.
+// Returns Addresses if non-empty; otherwise wraps Address in a slice.
+func (m *ModelUpstream) AllAddresses() []string {
+	if len(m.Addresses) > 0 {
+		return m.Addresses
+	}
+	if m.Address != "" {
+		return []string{m.Address}
+	}
+	return nil
+}
+
 // Load reads configuration from environment variables with defaults.
 func Load() (*Config, error) {
 	listenAddr := "127.0.0.1:8080"