Fix TextInfo.ToTitleCase Dutch IJ digraph and U+2019 apostrophe handling

[tarekgh]

This PR fixes two independent TextInfo.ToTitleCase bugs.

1. Dutch IJ digraph for the neutral nl culture

TextInfo.ToTitleCase special-cases Dutch to capitalize the ij digraph to IJ. The check gated this on CultureName.StartsWith("nl-"), which matches specific cultures (nl-NL, nl-BE, ...) but never the neutral Dutch culture, whose name is exactly "nl" with no trailing -. As a result the neutral culture fell through to the generic path:

CultureInfo.GetCultureInfo("nl-NL").TextInfo.ToTitleCase("de ijsvogel"); // "De IJsvogel"
CultureInfo.GetCultureInfo("nl").TextInfo.ToTitleCase("de ijsvogel");    // "De Ijsvogel"  (wrong)

Fix

Match the neutral name exactly or the nl- prefix:

string cultureName = CultureName;
bool isDutchCulture = cultureName.StartsWith("nl", StringComparison.OrdinalIgnoreCase) &&
    (cultureName.Length == 2 || cultureName[2] == '-');

The ij digraph rule is a property of the Dutch language, not of a specific region, so every nl* culture (neutral and specific alike) should apply it.

Tests

Added neutral nl cases to the existing DutchTitleCaseInfo_TestData theory in TextInfoTests, including the issue's repro ("de ijsvogel" -> "De IJsvogel").

Fixes #133032.

2. U+2019 (and related quotes) treated as apostrophes

ToTitleCase only treated the ASCII apostrophe ' (U+0027) as an in-word character. The typographic curly apostrophe ’ (U+2019) and similar characters have punctuation categories that are classified as word separators, so the letter after them was treated as the start of a new word:

CultureInfo.GetCultureInfo("en").TextInfo.ToTitleCase("Grandma’s pictures");
// before: "Grandma’S Pictures"   after: "Grandma’s Pictures"

Fix

Introduce an IsApostrophe helper and use it in the word scan. It covers the ASCII apostrophe plus the characters that share Unicode's Word_Break=MidNumLet property and are used as apostrophes:

private static bool IsApostrophe(char c)
{
    return c is '\'' or '\u2019' or '\u2018' or '\uFF07';
}

• U+2019 RIGHT SINGLE QUOTATION MARK (the curly apostrophe)
• U+2018 LEFT SINGLE QUOTATION MARK
• U+FF07 FULLWIDTH APOSTROPHE

Modifier-letter apostrophes such as U+02BC already work because their category (Lm) is not a word separator. Dot/colon-style MidLetter joiners (., :, ·, ...) are intentionally left out of scope to avoid changing titlecasing of dotted tokens.

Tests

Added a ToTitleCase_Apostrophe theory covering U+0027, U+2019, U+2018, and U+FF07, plus a control case ("Grandma-s" -> "Grandma-S") confirming genuine separators still end the word.

Fixes #133034.

Performance

No impact. The Dutch check remains a single StartsWith over a 2-character literal plus one length/index check, evaluated once per call. IsApostrophe is a small constant-time comparison replacing the previous single-character equality in the same loop position.

Note

This PR description was drafted with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟢 Approval recommended

Pull request overview

Fixes Dutch TextInfo.ToTitleCase handling for the neutral Dutch culture ("nl") by treating it as Dutch for the IJ digraph titlecasing rule, aligning behavior with specific Dutch cultures like "nl-NL".

Changes:

• Update the Dutch-culture detection in TextInfo.ToTitleCase to match "nl" and "nl-*" (case-insensitive).
• Extend existing Dutch titlecasing test data to cover the neutral "nl" culture, including a repro-style phrase case.

File summaries

File	Description
src/libraries/System.Private.CoreLib/src/System/Globalization/TextInfo.cs	Broadens the Dutch-culture gate so the IJ digraph rule applies to neutral "nl" in addition to "nl-*" cultures.
src/libraries/System.Runtime/tests/System.Globalization.Tests/System/Globalization/TextInfoTests.cs	Adds neutral "nl" cases to existing Dutch titlecasing theory data, covering both token and sentence inputs.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 0
• Review effort level: Lite

[Copilot]

🟢 Approval recommended

The behavioral changes are narrowly scoped, match the intended scenarios, and are covered by new targeted tests.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 1
• Review effort level: Lite