feat: M3.8.2 cache aligner integration — metrics + headers (3 tests)

CacheMetrics struct (40 LOC):
- stable_prefix_bytes, dynamic_tail_bytes
- drift_metric (0.0-1.0 ratio)
- cache_eligible flag (drift < 0.3)
- compression_ratio() and header_* methods

PromptBuilder::cache_metrics() (40 LOC):
- Calculates cache alignment metrics for query+chunk pairs
- Integrates CacheAligner output
- Gets compression ratio from ContextOptimizer
- Used for HTTP headers and observability

HTTP headers ready for client integration:
- X-Cache-Stable-Bytes
- X-Cache-Drift
- X-Cache-Eligible
- X-Compression-Ratio

3 new tests:
- test_cache_metrics_stable_query
- test_cache_metrics_compression_ratio
- test_cache_metrics_header_drift

Total: 117 mem-core tests (114 before + 3 new)

tasks/M3.8.2-cache-aligner-headers.md

@@ -0,0 +1,81 @@
+# M3.8.2 — CacheAligner: HTTP Headers + Metrics
+
+| Field | Value |
+|---|---|
+| Phase | M3.8 — Context optimization |
+| Size | M — 1–2 days |
+| Status | ⬜ Not started |
+| Depends | M3.8.1 (CacheAligner complete) |
+| Blocks | M3.8.3 |
+
+## Goal
+
+Integrate CacheAligner output into HTTP response headers and observability metrics
+so that:
+1. LLM provider can use cache hints
+2. Monitoring can track cache effectiveness
+3. Debugging can identify cache misses
+
+## Deliverables
+
+### 1. PromptBuilder::cache_metrics()
+
+New method returning cache metadata:
+
+```rust
+pub struct CacheMetrics {
+    pub stable_prefix_bytes: usize,
+    pub dynamic_tail_bytes: usize,
+    pub drift_metric: f32,  // 0.0-1.0 ratio
+    pub cache_eligible: bool,  // true if drift < 0.3
+}
+
+impl PromptBuilder {
+    pub fn cache_metrics(query: &Query, chunk: &Chunk) -> Result<CacheMetrics>
+}
+```
+
+Tests (3):
+- `test_cache_metrics_stable_query`
+- `test_cache_metrics_high_drift`
+- `test_cache_metrics_zero_drift`
+
+### 2. HTTP Response Headers
+
+Add to PromptBuilder output:
+- `X-Cache-Stable-Bytes`: size of cacheable prefix
+- `X-Cache-Drift`: 0.0-1.0 ratio
+- `X-Cache-Eligible`: "true"/"false"
+- `X-Compression-Ratio`: original vs. compressed
+
+Tests (4):
+- `test_headers_present_in_response`
+- `test_headers_accurate_values`
+- `test_headers_skipped_when_disabled`
+- `test_headers_format_valid`
+
+### 3. Observability Hooks
+
+Integrate with logging:
+```rust
+pub fn log_cache_metrics(metrics: &CacheMetrics) {
+    tracing::info!(
+        stable_bytes = metrics.stable_prefix_bytes,
+        drift = metrics.drift_metric,
+        eligible = metrics.cache_eligible,
+        "cache_alignment"
+    );
+}
+```
+
+Tests (2):
+- `test_metrics_logged_on_alignment`
+- `test_drift_high_triggers_warning`
+
+## Acceptance
+
+- All 9 new tests passing
+- Existing 114 mem-core tests still pass
+- Cache metrics accurately reflect alignment
+- HTTP headers present and valid
+- Zero performance overhead (< 1ms additional)