gh-127550 Fix ElementTree write newline handling

[durawat]

gh-127550: Fix ElementTree.write to ensure newline consistency between standalone and context manager use

Summary

Ensure consistent newline handling in ElementTree.write() when writing
to a filename versus a file-like object.

Relevant code in Lib/xml/etree/ElementTree.py

Function: _get_writer(file_or_filename, encoding)

Removing newline="\n" from the first code block caused multiple test
failures in Lib/test/test_xml_etree.py. In contrast, adding
newline="\n" to the second block resulted in only one failing test.

Writer for file-like objects

file = io.TextIOWrapper(
    file,
    encoding=encoding,
    errors="xmlcharrefreplace",
    newline="\n",
)

Writer for file or filename

   with open(file_or_filename, "w", encoding=encoding,
                  errors="xmlcharrefreplace", newline="\n") as file:

Changes

Added newline="\n" to the open() call for filename-based writing,
instead of removing it from the TextIOWrapper path, to ensure consistent
behavior between file-like objects and filenames.

This change improves consistency between the two code paths.

Rationale

According to the XML specification, XML parsers normalize line endings.
The sequences \r\n, \r, and \n are all normalized to \n during parsing.
Therefore, differences in newline handling at write time do not affect
parsed XML content, but consistency in output behavior is desirable.

Tests

• Fixed one failing test case in Lib/test/test_xml_etree.py
• Added a new regression test in Lib/test/test_xml_etree.py

[python-cla-bot]

The following commit authors need to sign the Contributor License Agreement:

• 123456+durawat@users.noreply.github.com

[bedevere-app]

Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool.

If this change has little impact on Python users, wait for a maintainer to apply the skip news label instead.

[bedevere-app]

Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool.

If this change has little impact on Python users, wait for a maintainer to apply the skip news label instead.

[picnixz]

The existing tests seem to indicate that this is a deliberate behavior. I am closing this PR because we need consensus first. Please don't open PRs in the future if no consensus is reached. In addition, don't use LLMs as we forbid them (read https://devguide.python.org/getting-started/generative-ai/ for acceptable uses)