feat(usage): track cached tokens + correct input/output/cache cost (#2209)

Normalize every provider to one cache-inclusive convention via
canonicalizeUsage() before persist, and price cached + cache_creation as
subsets of prompt_tokens in calculateCostFromTokens() to stop
double-counting. usageRepo now delegates cost math to a single source.
Surface Cached tokens/cost across dashboard (overview, tokens, cost,
details). Merge Claude message_start cache with message_delta output so
cache counts survive. Compatible LLM nodes now allow multiple API-key
connections (key pool).

Co-authored-by: Cursor <cursoragent@cursor.com>

src/app/api/providers/route.js

@@ -126,6 +126,8 @@ export async function POST(request) {
 
     let providerSpecificData = normalizeProviderSpecificData(provider, body, body.providerSpecificData);
 
+    // Compatible LLM nodes support multiple API-key connections (key pool); runtime
+    // rotates/fails over via getProviderCredentials. Embedding nodes stay single-connection.
     if (isOpenAICompatibleProvider(provider)) {
       const node = await getProviderNodeById(provider);
       if (!node) {