feat: add issue_usage tracking and metrics for AI ticket enhancement and planning

Add `issue_usage` table with columns for user_uid, tokens, cost, and latency metrics, including a unique index on user_uid. Wire `accumulate_usage` into `enhance_ticket` and `generate_plan` to capture per-request AI usage, persist totals via `add_issue_usage` in both `IssueCreateService` and `PlanningReportService`, and expose aggregated usage as metric cards through `IssueTrackerService.collect_metrics`. Update planning API docs summary to reflect the new phased implementation document format.

devplacepy/services/openai_gateway/usage.py

@@ -284,6 +284,45 @@ def parse_usage_headers(headers) -> Optional[dict]:
     }
 
 
+USAGE_FIELDS = (
+    "calls",
+    "prompt_tokens",
+    "completion_tokens",
+    "total_tokens",
+    "cost_usd",
+    "upstream_latency_ms",
+    "total_latency_ms",
+)
+
+
+def new_usage_totals() -> dict:
+    return {field: 0 for field in USAGE_FIELDS}
+
+
+def accumulate_usage(totals: Optional[dict], response) -> None:
+    if totals is None:
+        return
+    parsed = parse_usage_headers(getattr(response, "headers", None))
+    if not parsed:
+        return
+    for field in USAGE_FIELDS:
+        totals[field] += parsed[field]
+
+
+def usage_metric_cards(usage: dict) -> list[dict]:
+    return [
+        {"label": "AI calls", "value": usage["calls"]},
+        {"label": "Total tokens", "value": usage["total_tokens"]},
+        {"label": "Prompt tokens", "value": usage["prompt_tokens"]},
+        {"label": "Completion tokens", "value": usage["completion_tokens"]},
+        {"label": "Total cost", "value": f"${usage['cost_usd']:.4f}"},
+        {"label": "Avg tokens/call", "value": usage["avg_tokens"]},
+        {"label": "Avg cost/call", "value": f"${usage['avg_cost_usd']:.6f}"},
+        {"label": "Avg latency", "value": f"{usage['avg_upstream_latency_ms']:.0f}ms"},
+        {"label": "Avg tokens/sec", "value": usage["avg_tokens_per_second"]},
+    ]
+
+
 class GatewayUsageLedger:
     def record(self, raw: dict, pricing: Pricing, context_map: dict) -> Optional[dict]:
         try: