Em dash removal, nav/form/terms updates

- Replace all em dashes with hyphens site-wide (22 files)
- Add Safety to nav
- Remove Express Interest button from enterprise header (form is the path now)
- Enterprise form: secondary headcount selection shows honest odds message; yes disables submit button
- Terms §5: add network layer indemnification (patent pending 64/036,821); no proprietary names
- Terms §4: Neuron Inference is live now, not Q3 2026

src/inference.el

@@ -1,4 +1,4 @@
-// components/inference.el — Neuron inference section.
+// components/inference.el - Neuron inference section.
 // "My model. My infrastructure. Cheaper than theirs."
 
 fn inference() -> String {
@@ -10,20 +10,20 @@ fn inference() -> String {
       <div>
         <div class=\"inference-label-row reveal\">
           <div class=\"navy-line-left\" style=\"width:3rem;flex-shrink:0\"></div>
-          <span class=\"label\">Neuron inference &#8212; live day one</span>
+          <span class=\"label\">Neuron inference - live day one</span>
         </div>
         <h2 class=\"display-lg inference-headline reveal\" style=\"transition-delay:80ms\">
           My model. My infrastructure.
           <span class=\"gold\"> Cheaper than theirs.</span>
         </h2>
         <p class=\"inference-body reveal\" style=\"transition-delay:160ms\">
-          Four companies control nearly all frontier AI inference today. Every query you send strengthens their position &#8212; and their pricing power over you. That changes today.
+          Four companies control nearly all frontier AI inference today. Every query you send strengthens their position - and their pricing power over you. That changes today.
         </p>
         <p class=\"inference-body reveal\" style=\"transition-delay:240ms\">
           Neuron is both the product and the model. Purpose-built for how Neuron thinks, how your context is structured, how to do more with fewer tokens. It runs on Soma, our own inference infrastructure. Priced below OpenAI, Anthropic, and Google from day one. Not as a loss leader. As a permanent competitive position.
         </p>
         <p class=\"inference-body reveal\" style=\"transition-delay:320ms\">
-          You can still route to your own API keys if you want &#8212; Anthropic, OpenAI, whoever. But when we&#39;re cheaper, faster, and the model knows Neuron&#39;s architecture from the inside, why would you?
+          You can still route to your own API keys if you want - Anthropic, OpenAI, whoever. But when we&#39;re cheaper, faster, and the model knows Neuron&#39;s architecture from the inside, why would you?
         </p>
         <p class=\"inference-body reveal\" style=\"transition-delay:400ms\">
           Every user on Neuron inference is compute that doesn&#39;t flow to the monopoly. That&#39;s not just a cost story. It&#39;s a power one.
@@ -41,16 +41,16 @@ fn inference() -> String {
         <div class=\"stat-row reveal\" style=\"transition-delay:120ms\">
           <div class=\"stat-row-head\">
             <span class=\"stat-num\">Neuron</span>
-            <span class=\"stat-label\">The model &#8212; live on day one</span>
+            <span class=\"stat-label\">The model - live on day one</span>
           </div>
-          <p class=\"stat-body\">Built on Qwen3 &#8212; open weights, Sonnet-equivalent capability. Neuron wraps it with your full context, your memory graph, and years of accumulated intelligence. The model is the floor. Neuron is the ceiling.</p>
+          <p class=\"stat-body\">Built on Qwen3 - open weights, Sonnet-equivalent capability. Neuron wraps it with your full context, your memory graph, and years of accumulated intelligence. The model is the floor. Neuron is the ceiling.</p>
         </div>
         <div class=\"stat-row reveal\" style=\"transition-delay:240ms\">
           <div class=\"stat-row-head\">
             <span class=\"stat-num\">&lt; theirs</span>
             <span class=\"stat-label\">Priced below the incumbents</span>
           </div>
-          <p class=\"stat-body\">Below OpenAI&#39;s API pricing at launch. Below Anthropic&#39;s. Below Google&#39;s. Metered &#8212; you pay for what you use, not a flat subscription. And cheaper than any of them on every tier.</p>
+          <p class=\"stat-body\">Below OpenAI&#39;s API pricing at launch. Below Anthropic&#39;s. Below Google&#39;s. Metered - you pay for what you use, not a flat subscription. And cheaper than any of them on every tier.</p>
         </div>
       </div>