refactor: repartition server-side and client-side code

app/.server/llm/constants.ts

@@ -0,0 +1,7 @@
+// see https://docs.anthropic.com/en/docs/about-claude/models
+export const MAX_TOKENS = process.env.MAX_TOKENS ? parseInt(process.env.MAX_TOKENS) : 8000;
+
+// limits the number of model responses that can be returned in a single request
+export const MAX_RESPONSE_SEGMENTS = process.env.MAX_RESPONSE_SEGMENTS
+  ? parseInt(process.env.MAX_RESPONSE_SEGMENTS)
+  : 5;