Fix: exclude temperature parameter for reasoning-tier models

Anthropic has deprecated the temperature parameter for their reasoning
models (claude-opus-4-7). This was causing a 400 error when analyzeGraph
called callLLM with tier: 'reasoning'.

Solution: conditionally exclude temperature from the request body when
tier === 'reasoning'. Fast and standard tiers retain their temperature
parameter.

This unblocks the "Analyse and Optimize" button in the Knowledge Graph
admin panel post-Phase-2 deployment.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/lib/llm.js

@@ -303,9 +303,12 @@ export async function callLLM(options) {
   const body = {
     model,
     max_tokens: maxTokens,
-    temperature,
     messages: messagesPayload,
   };
+  // Temperature is not supported for reasoning tier models
+  if (tier !== 'reasoning') {
+    body.temperature = temperature;
+  }
   if (system !== undefined) body.system = system;
   if (tools && tools.length) body.tools = tools;
   if (toolChoice) body.tool_choice = toolChoice;