fix(index): skip indented comment lines in WebSource url files

[longzhenren]

Root cause

WebSource.__init__ parsed url files with:

if line.strip() and not line.startswith("#")

The emptiness check used the stripped line but the comment check used the raw line, so a comment with leading whitespace or a tab was treated as a URL and entered the render/index pipeline.

Repro on main:

# urls file: "https://a.com\n  # indented comment\n"
# before: ["https://a.com", "# indented comment"]
# after:  ["https://a.com"]

Fix

Strip each line once and apply both checks to the cleaned value (walrus in the comprehension).

Test

Added tests/test_web_source_urls.py covering plain URLs, indented/tab comment lines, blank and full-line comments, and whitespace stripping.

[vercel]

@longzhenren is attempting to deploy a commit to the andylizf's projects Team on Vercel.

A member of the Team first needs to authorize it.