feat: implement separate diagnostics for variables named '_' and unnecessary typed discards

[Copilot]

This PR addresses feedback from #888 by implementing separate diagnostic IDs and improving the analyzer's precision for different scenarios involving underscores and discards.

Changes Made

Separate Diagnostic IDs

• PH2147: "Avoid variables named ''" - flags variables explicitly named "" that can be confused with discards
• PH2152: "Avoid unnecessary typed discard" - flags typed discards like out string _ when out _ would suffice

Enhanced Overload Resolution Detection

Implemented simplified but effective overload detection that accurately determines when typed discards are necessary:

// This will be flagged - typed discard unnecessary
GetValue(out string _);  // Should use: out _

// This will NOT be flagged - typed discard necessary for overload resolution  
Parse("123", out int _);    // Disambiguates from Parse(string, out string)
Parse("test", out string _); // Disambiguates from Parse(string, out int)

Named Arguments Support

Added proper handling for named arguments in overload resolution:

// Both positional and named arguments work correctly
GetValue(value: out string _);  // Flagged as unnecessary
Parse(input: "123", result: out int _);  // Not flagged if needed for overload resolution

Documentation and Performance Improvements

• Split documentation: Created separate documentation files for PH2147 and PH2152
• Performance optimization: Moved SemanticModel access lower in method for better performance
• Simplified LINQ: Removed unnecessary operations as suggested in code review
• Fixed diagnostic ID conflict: Resolved PH2153 mapping conflict between analyzers

Code Quality Improvements

• Resolved duplicate code detection issues (PH2071) in test methods
• Uses DiagnosticAnalyzerBase for proper dual diagnostic support
• Maintains all original functionality for variables named "_"
• Improved test coverage with all 15 tests passing

Impact

• More precise diagnostics: Separate IDs allow different handling of variable naming vs discard usage
• Better overload detection: Uses semantic analysis to detect when typed discards are truly necessary
• Named argument support: Handles modern C# syntax correctly
• Cleaner code: Follows analyzer best practices and formatting standards
• Better performance: Optimized semantic model usage

This maintains the original intent of promoting anonymous discards (out _) while ensuring typed discards are only flagged when truly unnecessary for overload resolution.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.