Fix SeparatePanel fullscreen origin by clearing WA_Moved

[anriltine]

Summary

• Clear Qt::WA_Moved on SeparatePanel before showFullScreen(), so Qt can place the frameless Mini App panel at the screen origin instead of only resizing it.
• Follows the Linux transient-parent path in initGeometry(), which already clears WA_Moved after RpWidget's constructor setGeometry(0, 0, 0, 0).

Context

Reported in telegramdesktop/tdesktop#30963 (also #31041 / #31114).

This supersedes #342. That PR forced screen geometry after showFullScreen() and blamed QTBUG-39537 / QTBUG-86899. @ilya-fedin pointed out those bugs are unrelated/fixed, and that the real issue is likely ours: Qt only auto-positions unless setGeometry() / move() were called, so WA_Moved from RpWidget keeps Mini Apps anchored at the pre-fullscreen origin.

Forcing SetGeometryAndScreen() after showFullScreen() is also the wrong tool: setGeometry_sys() clears Qt::WindowFullScreen when not inside setWindowState().

Test plan

• Windows, single monitor: open a Mini App, enter fullscreen via UI or web_app_request_fullscreen — panel should cover the full screen from the monitor origin.
• Windows, two monitors: main Telegram window mid-primary; open Mini App fullscreen — panel should cover the primary screen, not remain centered over the main window and overflow.
• Open a Mini App that requests fullscreen immediately — same full-screen origin.
• Exit fullscreen — panel returns to the normal anchored size/position.
• Linux/macOS smoke: Mini App fullscreen still enters/exits correctly.

Made with Cursor

[anriltine]

@ilya-fedin this is the clean follow-up to #342 using setAttribute(Qt::WA_Moved, false) before showFullScreen(), as you suggested.

[ilya-fedin]

You checked it works, right?

[anriltine]

No, included the test plan)

[ilya-fedin]

Please tell when you have checked it works