Converting an XMLText field to RichText can produce a validation error, and migrated RichText field will throw "Content did not validate" exceptions later, making it uneditable.

Steps to reproduce
1. Clean installation of eZ Platform 2.5/Ibexa DXP 3.x
2. Install https://github.com/ezsystems/ezplatform-xmltext-fieldtype
3. Create a new Content Type - XML Test with fields: 

• name (single line)
• text (XMLBlock)
  4. Create a new Content of type XML Test with text field of value:

  <?xml version="1.0" encoding="UTF-8"?>
  <section xmlns:custom="http://ez.no/namespaces/ezpublish3/custom/" xmlns:image="http://ez.no/namespaces/ezpublish3/image/" xmlns:xhtml="http://ez.no/namespaces/ezpublish3/xhtml/">
     <section>
        <paragraph xmlns:tmp="http://ez.no/namespaces/ezpublish3/temporary/">
           <table border="0" width="400px">
              <tr>
                 <td xhtml:width="100"/>
                 <td xhtml:width="150">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
                 <td xhtml:width="150">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
              </tr>
              <tr>
                 <td xhtml:width="100">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
                 <td xhtml:width="150">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
                 <td xhtml:rowspan="3" xhtml:width="150" custom:valign="center">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
              </tr>
              <tr>
                 <td xhtml:width="100">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
                 <td xhtml:width="150">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
              </tr>
              <tr>
                 <td xhtml:width="100">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
                 <td xhtml:width="150">
                    <paragraph>Lorem ipsum</paragraph>
                 </td>
              </tr>
           </table>
        </paragraph>
     </section>
  </section>

  5. Publish it.
  6. Clear cache and run conversion command:

  php bin/console ezxmltext:convert-to-richtext 

Result
The conversion command will report:

ERROR     [app] Validation errors when converting ezxmltext for contentobject_attribute.id=X 

Expected result
The conversion command completes without errors.

Note
The underlying error is:

Error in 2:0: Element section has extra content: informaltable 

XML output for RichText:

<?xml version="1.0" encoding="UTF-8"?>
<section xmlns="http://docbook.org/ns/docbook" xmlns:ezcustom="http://ez.no/xmlns/ezpublish/docbook/custom" xmlns:ezxhtml="http://ez.no/xmlns/ezpublish/docbook/xhtml" xmlns:xlink="http://www.w3.org/1999/xlink" version="5.0-variant ezpublish-1.0">
   <informaltable width="400">
      <tbody>
         <tr>
            <td ezxhtml:width="100" />
            <td ezxhtml:width="150">
               <para>Lorem ipsum</para>
            </td>
            <td ezxhtml:width="150">
               <para>Lorem ipsum</para>
            </td>
         </tr>
         <tr>
            <td ezxhtml:width="100">
               <para>Lorem ipsum</para>
            </td>
            <td ezxhtml:width="150">
               <para>Lorem ipsum</para>
            </td>
            <td ezxhtml:width="150" valign="center" rowspan="3">
               <ezattribute>
                  <ezvalue key="valign">center</ezvalue>
               </ezattribute>
               <para>Lorem ipsum</para>
            </td>
         </tr>
         <tr>
            <td ezxhtml:width="100">
               <para>Lorem ipsum</para>
            </td>
            <td ezxhtml:width="150">
               <para>Lorem ipsum</para>
            </td>
         </tr>
         <tr>
            <td ezxhtml:width="100">
               <para>Lorem ipsum</para>
            </td>
            <td ezxhtml:width="150">
               <para>Lorem ipsum</para>
            </td>
         </tr>
      </tbody>
   </informaltable>
</section>

During tests, I was able to spot Errors for different tags as well, for example, title.