docs: Reword YOLOv8 bullet to mention Roboflow.

Drop the marketing-y "three lines of glue" claim and link to Roboflow
as the suggested place to train a custom YOLOv8 model.

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

Author: kwagyeman

docs/conf.py

@@ -242,7 +242,6 @@ def _render_landing_code(src):
     k: _render_landing_code(v) for k, v in _landing_examples_src.items()
 }
 
-
 # -- General configuration ------------------------------------------------
 
 # If your documentation needs a minimal Sphinx version, state it here.
@@ -569,7 +568,6 @@ def _render_landing_code(src):
 # If true, show URL addresses after external links.
 # man_show_urls = False
 
-
 # -- Options for Texinfo output -------------------------------------------
 
 # Grouping the document tree into Texinfo files. List of tuples
@@ -599,6 +597,5 @@ def _render_landing_code(src):
 # If true, do not generate a @detailmenu in the "Top" node's menu.
 # texinfo_no_detailmenu = False
 
-
 # Example configuration for intersphinx: refer to the Python standard library.
 intersphinx_mapping = {"python": ("https://docs.python.org/3", None)}

docs/templates/topindex.html

@@ -819,7 +819,7 @@ <h3>Real-time person tracking</h3>
             <p>The on-board YOLOv8 model is a single-class person detector — int8 quantised and shipped in ROM.</p>
             <div class="omv-bullet"><span class="check"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="3" stroke-linecap="round"><path d="M5 12l5 5L20 7"/></svg></span><span>Loaded from <code>/rom/yolov8n_192.tflite</code> — no SD card or download needed.</span></div>
             <div class="omv-bullet"><span class="check"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="3" stroke-linecap="round"><path d="M5 12l5 5L20 7"/></svg></span><span>Runs in real time on NPU-equipped boards — the OpenMV N6 and AE3.</span></div>
-            <div class="omv-bullet"><span class="check"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="3" stroke-linecap="round"><path d="M5 12l5 5L20 7"/></svg></span><span>Train your own YOLOv8 in Ultralytics — same three lines of glue.</span></div>
+            <div class="omv-bullet"><span class="check"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="3" stroke-linecap="round"><path d="M5 12l5 5L20 7"/></svg></span><span>Bring your own YOLOv8 model trained on <a href="https://roboflow.com">Roboflow</a> and load it the same way.</span></div>
           </div>
         </div>
         <div class="tab-body" data-tab="apriltag" role="tabpanel">