docs: move context-menu screenshot into Usage and shrink to ~1/3 width

Author: max3925vats

README.md

@@ -18,8 +18,6 @@ Markdown is plain text, so converted papers drop straight into note apps like
 Obsidian, AI/RAG pipelines, citation extraction, summarisation, literature
 reviews — anything that prefers text over PDFs.
 
-![Right-click a PDF attachment and choose "Convert with Docling"](docs/images/context-menu.png)
-
 ---
 
 ## Quick start
@@ -222,6 +220,9 @@ Then install the `.xpi` via **Tools → Plugins** as above.
 4. (Optional) tick **Auto-convert new PDF attachments** in Behavior to run
    conversion automatically as you import new PDFs.
 
+<img src="docs/images/context-menu.png" alt="Right-click a PDF attachment and choose &quot;Convert with Docling&quot;" width="270">
+
+
 All docling-serve options exposed in the preferences pane (pipeline, OCR,
 VLM presets, enrichments, etc.) live inside two collapsible disclosure
 sections — **Conversion options** and **Advanced** — collapsed by default so