Add Serialization Benchmark

[corydolphin]

Introduce benchmarks using a large (~117KB) GraphQL query to measure
parse and pickle serialization performance. These provide a baseline
for comparing serialization approaches in subsequent commits.

Baseline performance (measrued on a Macbook Pro M4 Max):

• Parse: 81ms
• Pickle encode: 24ms
• Pickle decode: 42ms
• Roundtrip: 71ms

Thank you for contributing to GraphQL-core!

If your pull-request is non-trivial, adds a feature or contains a non-breaking change, then please, first open an issue to discuss the proposed changes and add a link to that issue.

GraphQL-core tries very hard to stay within the scope of being just a Python port of GraphQL.js.

Any additional feature or incompatible change will be only accepted in rare cases and requires a compelling reason, because they aggravate maintenance and synchronization with the developments in the upstream project. So please discuss such changes upfront in an issue before sending a PR. Maybe there are other ways to solve the problem.

If possible, also add unit tests, or provide runnable example code as part of the accompanying issue, from which unit tests can be derived.

[corydolphin]

This change is part of the following stack:

• Add Serialization Benchmark #8 ◀
  • Use tuples for all AST collection fields (instead of lists) #9
    • Update minimum Python version to 3.10 #10
      • Make visitor immutable-friendly #11
        • Fix tests to use proper AST node construction and types #12
          • Fix parser type safety for stricter AST types #13
            • Convert AST nodes to frozen dataclasses (70% faster decode, 40% faster parsing) #14

_{Change managed by git-spice.}