fix: address markdown lint and doc accuracy issues (threads 6-13)

Thread 6: Add python language specifier to .claude-plugin/README.md code fence
Thread 7: Fix MD022/MD031 spacing in docs/MARKETPLACE.md
Thread 8: Blank line before Destructive Actions heading in README.md
Thread 9: Move H1 to first line in skills/unraid/references/api-reference.md
Thread 10+11+12: Rewrite quick-reference.md:
  - Fix MD022/MD031 (blank lines around headings/fences)
  - Add python language specifiers to all code fences
  - Fix disk/logs params: path/lines → log_path/tail_lines
  - Add path param to live/log_tail example
  - Remove invalid notification/unread subaction
  - Fix notification/list filter= → list_type=
Thread 13: Add python/text language specifiers to troubleshooting.md fences
Thread 14: Update test-tools.sh header comment (in separate commit)
Also: Remove 'connect' from README system subactions table (it was removed from
the tool in previous commit), fix notification 'unread' → 'mark_unread' in table

skills/unraid/references/troubleshooting.md

@@ -5,7 +5,8 @@
 **Error:** `CredentialsNotConfiguredError` or message containing `~/.unraid-mcp/.env`
 
 **Fix:** Run setup to configure credentials interactively:
-```
+
+```python
 unraid(action="health", subaction="setup")
 ```
 
@@ -20,12 +21,14 @@ This writes `UNRAID_API_URL` and `UNRAID_API_KEY` to `~/.unraid-mcp/.env`. Re-ru
 **Diagnostic steps:**
 
 1. Test basic connectivity:
-```
+
+```python
 unraid(action="health", subaction="test_connection")
 ```
 
 2. Full diagnostic report:
-```
+
+```python
 unraid(action="health", subaction="diagnose")
 ```
 
@@ -57,7 +60,8 @@ unraid(action="health", subaction="diagnose")
 **Error:** `Action 'X' was not confirmed. Re-run with confirm=True to bypass elicitation.`
 
 **Fix:** Add `confirm=True` to the call:
-```
+
+```python
 unraid(action="array", subaction="stop_array", confirm=True)
 unraid(action="vm",    subaction="force_stop", vm_id="<id>", confirm=True)
 ```