feat: antropik cache — default anthropic-version, forward cache headers, model features, streaming usage, tests

- Auto-default anthropic-version to 2023-06-01 for prompt caching
- Forward cache headers (x-cache, cf-cache-status, age, etc.) from backend
- Add features: ["prompt_caching"] to /v1/models response
- Extract and report actual usage tokens in streaming message_delta
- Track output char count for token estimation when backend omits it
- Cache-aware usage reporting: cache_creation_input_tokens, cache_read_input_tokens
- Comprehensive tests for cache_control preservation in content blocks & system

src/lib/router.ts

@@ -494,15 +494,17 @@ export async function handleRequest(
 	if (url.pathname === "/v1/models" && req.method === "GET") {
 		const authErr = requireAuth(req, env);
 		if (authErr) return authErr;
+		const models = listModels().map((id) => ({
+			id,
+			object: "model",
+			created: Math.floor(Date.now() / 1000),
+			owned_by: "edge-proxy",
+			features: ["prompt_caching"],
+		}));
 		return new Response(
 			JSON.stringify({
 				object: "list",
-				data: listModels().map((id) => ({
-					id,
-					object: "model",
-					created: Math.floor(Date.now() / 1000),
-					owned_by: "edge-proxy",
-				})),
+				data: models,
 			}),
 			{
 				status: 200,