Long post detection should probably not count any text following a ">" or inside a "``` " code block

feedback: It would be great if the displayed content could be markdown-ified into HTML (most critically the ">" reply blocks and code blocks``` being plain text makes it hard to find the actual person's text that we want to read)

The issue names no files, tests, or entry points. Start by locating the long-post detection and displayed-content rendering, then verify how quoted text beginning with ">" and fenced code blocks are handled; done means those sections no longer affect detection and displayed content is rendered as readable HTML.

Written by the indexing model from the issue text.