docs: fix DNS authentication to use root domain instead of _mcp-registry subdomain (#388)

Fixes #385

Updates DNS authentication documentation to correctly specify that TXT
records should be placed at the root domain (e.g., `example.com`) rather
than the `_mcp-registry` subdomain.

The implementation correctly looks up TXT records at the root domain,
but the documentation incorrectly instructed users to create records at
`_mcp-registry.<domain>`. This mismatch was causing authentication
failures.

## Why use root domain instead of subdomain?

The root domain approach is clearer: if you want to publish to
`com.example/my-server`, you authenticate with `example.com`, not
`_mcp-registry.example.com`.

It also aligns with other TXT DNS standards, e.g. SPF. (That said, there
are other DNS standards that use subdomains e.g. DKIM 🤷)

## Changes
- Updated publish-server.md to remove `_mcp-registry.` prefix from DNS
examples
- Updated CLI commands reference documentation  
- Updated namespacing explanation documentation

docs/guides/publishing/publish-server.md

@@ -302,7 +302,7 @@ This opens your browser for OAuth authentication.
 openssl genpkey -algorithm Ed25519 -out key.pem
 
 # Get public key for DNS record
-echo "_mcp-registry.yourcompany.com. IN TXT \"v=MCPv1; k=ed25519; p=$(openssl pkey -in key.pem -pubout -outform DER | tail -c 32 | base64)\""
+echo "yourcompany.com. IN TXT \"v=MCPv1; k=ed25519; p=$(openssl pkey -in key.pem -pubout -outform DER | tail -c 32 | base64)\""
 
 # Add the TXT record to your DNS, then login
 mcp-publisher login dns --domain yourcompany.com --private-key $(openssl pkey -in key.pem -noout -text | grep -A3 "priv:" | tail -n +2 | tr -d ' :\n')