docs: fix thundering herd framing — agents converge because of shared training data

Not "agents serve humans who ask the same questions" but "agents
trained on the same data independently reach the same conclusions
and make the same queries." The convergence is from training data,
not from humans.

Author: teamchong

docs/src/content/docs/why-querymode.mdx

@@ -3,7 +3,7 @@ title: Why QueryMode
 description: Agents query like humans, but at machine pace. Data infrastructure needs to keep up.
 ---
 
-1. **Agents are becoming the majority of internet traffic.** They serve different owners across different parts of the world, but they all serve humans — and humans ask the same questions. When thousands of agents hit the same endpoints at the same millisecond with the same human query, the result is thundering herds that look like a DDoS — except every request is legitimate. That's not a DDoS attack. That's just Tuesday. Data must live at the edge to survive this.
+1. **Agents are becoming the majority of internet traffic.** They're trained on the same data, so they reach the same conclusions and make the same decisions. When thousands of agents independently arrive at the same query — not because they copied each other, but because the training data led them there — the result is correlated traffic that looks like a DDoS. Except every request is legitimate. That's not an attack. That's just Tuesday. Data must live at the edge to survive this.
 
 2. **Agents need live data.** Decisions based on outdated training data lead to bad outcomes. Training data can't keep up with the speed the world produces information. Agents make API calls for live data — lots of them. That data needs to live at the edge, close to where agents run.