Optimize hot paths: cache NuGetFramework parsing, avoid string allocations in hash computation

[Copilot]

Identified and fixed performance bottlenecks in package version checking and hash computation paths.

Changes

• Cache parsed NuGetFramework instances (NuGetMetadataService.cs)
  • CompatibleTargetFrameworksTyped was re-parsing frameworks on every access
  • Now uses Lazy<T> for thread-safe caching

// Before: O(n) parsing on every property access
public IEnumerable<NuGetFramework> CompatibleTargetFrameworksTyped => 
    CompatibleTargetFrameworks.Select(tf => NuGetFramework.Parse(tf));

// After: parse once, cache result
private Lazy<IReadOnlyList<NuGetFramework>>? _compatibleTargetFrameworksTyped;
public IReadOnlyList<NuGetFramework> CompatibleTargetFrameworksTyped => 
    (_compatibleTargetFrameworksTyped ??= new Lazy<IReadOnlyList<NuGetFramework>>(() => 
        CompatibleTargetFrameworks.Select(tf => NuGetFramework.Parse(tf)).ToList())).Value;

• Use Count property instead of LINQ Count() (OutdatedService.cs)

  • Direct property access on HashSet<T> vs extension method

• Eliminate string allocations in hash computation (OutdatedService.cs)

  • Replaced str?.ToLowerInvariant().GetHashCode() with StringComparer.OrdinalIgnoreCase.GetHashCode(str)
  • Avoids allocating lowercase string copies per hash call

Original prompt

Identify and suggest improvements to slow or inefficient code

---

💡 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.