Fixed large json deserialization regression

[sunghwan2789]

While a rented array used to create a JsonDocument requires its lifecycle to be tied to the JsonDocument's, this PR introduces a new method that is free from the issues observed before.

Closes #8376

[Copilot]

Pull Request Overview

A memory-efficient fix for large JSON deserialization by replacing the pooled buffer approach with a streaming Pipe-based writer and reader.

• Replace PooledArrayWriter with Pipe and Utf8JsonWriter(pipe.Writer) in ResponseEnumerable
• Add a new integration test (TestLarge) and supporting LargeQueryResolvers
• Record a large-response snapshot for the new test

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
src/StrawberryShake/Client/src/Transport.Http/ResponseEnumerable.cs	Swap out in-memory buffer for Pipe streaming and update parsing call
AnyScalarDefaultSerializationTest.cs	Add Execute_AnyScalarDefaultSerialization_TestLarge test and LargeQueryResolvers
AnyScalarDefaultSerializationTest.Execute_AnyScalarDefaultSerialization_TestLarge.snap	New snapshot capturing large JSON output

Comments suppressed due to low confidence (1)

src/StrawberryShake/Client/src/Transport.Http/ResponseEnumerable.cs:91

• [nitpick] Switching to a Pipe-backed writer may introduce extra overhead for small payloads. Consider benchmarking or using a size threshold to fall back to the pooled buffer for smaller responses.

        var pipe = new Pipe();