gh-138071: Clarify curses.color_pair doc about style vs extraction masks

[PrinceNaroliya]

Documentation fix

This PR clarifies the wording in the curses.color_pair documentation.

Previously it said:

This attribute value can be combined with A_STANDOUT, A_REVERSE, and the other A_* attributes.

This was ambiguous because constants like A_COLOR, A_ATTRIBUTES, and A_CHARTEXT also follow the A_* pattern, but those are extraction masks and cannot be combined with curses.color_pair().

Changes made

• Updated Doc/library/curses.rst:
  • Explicitly clarified that the value returned by curses.color_pair(n) is an attribute mask that can be combined (using |) with style attributes like A_BOLD, A_REVERSE, A_UNDERLINE.
  • Added a note that A_COLOR, A_ATTRIBUTES, and A_CHARTEXT are extraction masks used with & on values returned from window.inch(), and must not be combined with color_pair().

Why this is needed

• Avoids confusion for new users who may assume all A_* constants can be OR'ed with curses.color_pair.
• Matches the behavior of curses in practice.

Example

# Valid
attr = curses.color_pair(3) | curses.A_BOLD | curses.A_REVERSE
win.addstr("Hello", attr)

# Invalid (no effect, misleading)
attr = curses.color_pair(3) | curses.A_COLOR


<!-- gh-issue-number: gh-138071 -->
* Issue: gh-138071
<!-- /gh-issue-number -->


<!-- readthedocs-preview cpython-previews start -->
----
📚 Documentation preview 📚: https://cpython-previews--138079.org.readthedocs.build/

<!-- readthedocs-preview cpython-previews end -->

[picnixz]

This is too verbose. I don't think we need to change this. A_* attributes are well-defined. A_COLOR is not categorized as an attribute in the corresponding table. Sorry, but I'm going to close this one.