fix(editor): database behavier (#14394)

fix #13459
fix #13707
fix #13924

#### PR Dependency Tree


* **PR #14394** 👈

This tree was auto-generated by
[Charcoal](https://github.com/danerwilliams/charcoal)

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **New Features**
* Improved URL paste: text is split into segments, inserted correctly,
and single-URL pastes create linked-page references.

* **UI Improvements**
  * Redesigned layout selector with compact dynamic options.
* Number-format options are always available in table headers and mobile
menus.

* **Bug Fixes**
  * More consistent paste behavior for mixed text+URL content.
  * Prevented recursive selection updates when exiting edit mode.

* **Tests**
* Added tests for URL splitting, paste insertion, number formatting, and
selection behavior.

* **Chores**
* Removed number-formatting feature flag; formatting now applied by
default.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

blocksuite/affine/shared/src/__tests__/utils/url.unit.spec.ts

@@ -1,6 +1,6 @@
 import { describe, expect, test } from 'vitest';
 
-import { isValidUrl } from '../../utils/url.js';
+import { isValidUrl, splitTextByUrl } from '../../utils/url.js';
 
 describe('isValidUrl: determining whether a URL is valid is very complicated', () => {
   test('basic case', () => {
@@ -85,3 +85,55 @@ describe('isValidUrl: determining whether a URL is valid is very complicated', (
     expect(isValidUrl('http://127.0.0.1', 'http://127.0.0.1')).toEqual(true);
   });
 });
+
+describe('splitTextByUrl', () => {
+  test('should split text and keep url part as link segment', () => {
+    expect(splitTextByUrl('hi - https://google.com')).toEqual([
+      { text: 'hi - ' },
+      {
+        text: 'https://google.com',
+        link: 'https://google.com',
+      },
+    ]);
+  });
+
+  test('should support prefixed url token without swallowing prefix text', () => {
+    expect(splitTextByUrl('-https://google.com')).toEqual([
+      { text: '-' },
+      {
+        text: 'https://google.com',
+        link: 'https://google.com',
+      },
+    ]);
+  });
+
+  test('should trim tail punctuations from url token', () => {
+    expect(splitTextByUrl('visit https://google.com, now')).toEqual([
+      { text: 'visit ' },
+      {
+        text: 'https://google.com',
+        link: 'https://google.com',
+      },
+      { text: ', now' },
+    ]);
+  });
+
+  test('should convert domain token in plain text', () => {
+    expect(splitTextByUrl('google.com and text')).toEqual([
+      {
+        text: 'google.com',
+        link: 'https://google.com',
+      },
+      { text: ' and text' },
+    ]);
+  });
+
+  test('should normalize www domain token link while preserving display text', () => {
+    expect(splitTextByUrl('www.google.com')).toEqual([
+      {
+        text: 'www.google.com',
+        link: 'https://www.google.com',
+      },
+    ]);
+  });
+});