[lexical-markdown] Bug Fix: Preserve markdown hard line breaks during import when newlines are preserved

[joshjryan]

Description

I noticed that hard line breaks in markdown could get lost after an import and edit flow, when the markdown is toggled into the editor and then saved back out.

(this is my first contribution to this project, so please let me know if I should be doing anything differently)

This PR keeps hard-break intent intact when newline preservation is enabled:

Handles two trailing spaces before a newline as a hard break marker during import line merge handling.

• Keeps existing backslash hard-break behavior aligned in the same path.
• Adds focused unit coverage for both hard-break forms so round-trip behavior is explicit:
  -- foo followed by two spaces and a newline
  -- foo followed by backslash and a newline

I also added coverage around normalization behavior so the merge path is tested directly.

Closes #4843

Test plan

Before

• Hard line breaks could be lost in markdown import/export round trips in newline-preserving workflows.
• No dedicated round-trip test for backslash hard breaks in that path.
• No dedicated test asserting two-space hard-break handling in that same path.

###After

• Added unit test coverage that round-trips both hard-break forms with newline preservation enabled:
  -- markdown hard line break import: preserves hard line break when shouldPreserveNewLines is true
  -- markdown hard line break import: preserves backslash hard line break when shouldPreserveNewLines is true
• Added normalization coverage for hard-break-adjacent merge behavior.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
lexical	Ready	Preview, Comment	Mar 28, 2026 9:40pm
lexical-playground	Ready	Preview, Comment	Mar 28, 2026 9:40pm

[etrepum]

This PR doesn't look like it solves the issue that it claims to close, this one only changes the behavior of trailing backslash and not leading/trailing whitespace

[joshjryan]

This PR doesn't look like it solves the issue that it claims to close, this one only changes the behavior of trailing backslash and not leading/trailing whitespace

You're right, I missed the part of the request about leading whitespace. I was just focused on trailing whitespace (for indicating hard line-breaks with trailing double spaces).

I can modify this to also account for leading white space.

[etrepum]

This seems reasonable at least as far as the existing test suite goes