Exercise 3: Two Sibling Paragraphs — Possible Solution
====================================================================

THE TEST
------------------------------
    html = "<div><p>One</p><p>Two</p></div>"
    builder = TinyDOMBuilder()
    builder.feed(html)
    print(builder.root)

HAND SKETCH (before running)
------------------------------
    #document
      div
        p
          "One"
        p
          "Two"

Both <p> elements should sit at the SAME indentation level, as two
separate children of the one div -- not nested inside each other,
since the source closes each <p> (</p>) before the next one opens.

ACTUAL VERIFIED OUTPUT
------------------------------
    Element('#document')
      Element('div')
        Element('p')
          Text('One')
        Element('p')
          Text('Two')

Matches the hand sketch exactly.

WHY THIS WORKS AS AN ANSWER
------------------------------
TinyDOMBuilder's own handle_endtag pops the stack the moment a
closing tag is seen -- so by the time the parser reaches the second
<p>, the first one has already been popped back off, leaving div (not
the first p) as the current top of the stack. The second p therefore
gets appended as div's own child, at the same level as the first p,
rather than being nested inside it. This is the exact mechanism that
makes "next to" and "nested inside" genuinely different outcomes in a
real DOM tree: it isn't about how the two elements LOOK when written
next to each other in the source -- it's about whether a closing tag
was seen (popping back up a level) before the next opening tag
arrived. Two sibling <p> elements close cleanly before the next one
starts; two NESTED elements (like this chapter's own <div><div>...
example) don't close until after the inner one has already opened,
which is exactly why the stack-based approach handles both cases
correctly without needing any special-casing for either shape.