Build a markdown-to-HTML converter in the sandbox called md2html.py.

It reads a small markdown document from standard input and prints the converted
HTML to standard output. Support exactly this subset:

- `# Heading`   -> `<h1>Heading</h1>`
- `## Heading`  -> `<h2>Heading</h2>`
- `**bold**`    -> `<strong>bold</strong>`
- `` `code` ``  -> `<code>code</code>`
- `- item`      -> `<ul><li>item</li></ul>` (one list per contiguous block)
- blank line    -> separates blocks
- other text    -> `<p>text</p>`

HTML-escape the content (e.g. `<` and `&` in the source text must become
`&lt;` and `&amp;`). The exact output format matters — the verifier compares
against golden files character-for-character.

Test it against the verification oracle before replying DONE.
When the goal is met and verified, reply exactly: DONE.
