homepage: rework LLM model cards to structured role/spec/use lines

Each LLM card now carries a details object (role/spec/use) rendered as a
bold pale-green role, a bright-green spec line, and a dim 'Use:' line.
app.js descHTML() falls back to the plain description for non-LLM cards;
new .desc-role/.desc-spec/.desc-use CSS + bare-mode reset.

Roles made neutral (no nicknames): frank-large = deep reasoning,
frank-medium = primary model, frank-small = fast utility model.

The /docs model guide was fixed — it had drifted to retired model info
(frank-large as Qwen3.8-Flash-Next 87.3 GiB/200k, frank-medium as FP8
~31 GiB) — now Qwen3.8-27B UD-Q8_K_XL 31.5 GiB/262k and MXFP4/W4A8
19.4 GiB, reordered large->medium->small. bats 30/30 + live render.

shared/homepage/docs.html

@@ -45,26 +45,26 @@
   <p>Three models serve the family. The homepage status cards show whether
   each one is up right now.</p>
 
-  <h2>frank-large — the big brain</h2>
+  <h2>frank-large — deep reasoning</h2>
   <ul>
-    <li><strong>Model:</strong> Qwen3.8-Flash-Next, UD-IQ4_XS (87.3 GiB).</li>
-    <li><strong>Capacity:</strong> 200k-token context, exactly one conversation at a time. A second waits — politely, but it waits.</li>
-    <li><strong>Best for:</strong> cybersecurity and debugging tasks. <em>Use requires Frank's permission.</em></li>
+    <li><strong>Model:</strong> Qwen3.8-27B (UD-Q8_K_XL, 31.5 GiB).</li>
+    <li><strong>Capacity:</strong> 262k-token context, exactly one conversation at a time. Text only.</li>
+    <li><strong>Best for:</strong> cybersecurity and debugging. <em>Use requires Frank's permission.</em></li>
     <li><strong>Not for:</strong> images. It cannot see them.</li>
   </ul>
 
-  <h2>frank-small — the eyes</h2>
+  <h2>frank-medium — the primary model</h2>
   <ul>
-    <li><strong>Model:</strong> Gemma4-12B QAT (Balanced, Q4_K_M, 7.4 GiB).</li>
-    <li><strong>Capacity:</strong> one million tokens of pooled context across four simultaneous conversations (up to 262k each).</li>
-    <li><strong>Best for:</strong> the fast, cheap utility model — agentic search, text/audio/image RAG, research subagents, language translation, roleplaying games, summarization, and anything with a picture (screenshots, photos, diagrams). Any small utility task that doesn't require much intelligence.</li>
+    <li><strong>Model:</strong> Qwen3.8-27B-Uncensored (MXFP4/W4A8, 19.4 GiB).</li>
+    <li><strong>Capacity:</strong> 262k-token context shared by up to eight conversations at once; sees images.</li>
+    <li><strong>Best for:</strong> software engineering, system administration, and instruction tasks. Beats Claude Sonnet 5 per Artificial Analysis.</li>
   </ul>
 
-  <h2>frank-medium — the 27B generalist</h2>
+  <h2>frank-small — the fast utility model</h2>
   <ul>
-    <li><strong>Model:</strong> Qwen3.8-27B-Uncensored, FP8 (~31 GiB).</li>
-    <li><strong>Capacity:</strong> 262k-token context shared by up to eight conversations at once. It can see pictures too; <code>frank-small</code> is the faster eye and holds four conversations.</li>
-    <li><strong>Best for:</strong> the primary intelligent model — it surpasses Claude Sonnet 5 per Artificial Analysis. Use it for software engineering, system administration, and instruction tasks.</li>
+    <li><strong>Model:</strong> Gemma4-12B QAT (Balanced, Q4_K_M, 7.4 GiB).</li>
+    <li><strong>Capacity:</strong> one million tokens of pooled context across four simultaneous conversations (up to 262k each); fastest at images.</li>
+    <li><strong>Best for:</strong> agentic search, RAG, research subagents, translation, roleplay, and summarization — any small task that doesn't need much intelligence, and anything with a picture (screenshots, photos, diagrams).</li>
   </ul>
 
   <p class="foot-note">Which one to pick? Small utility work, RAG, search, and