Use a faster JSON library

[ofek]

Motivation

We parse JSON from multiple endpoints, with each payload being many MiB in size. Some users in resource constrained environments are experiencing abnormally high CPU usage. A more efficient parser should ameliorate that.

Additional Notes

orjson is currently the fastest JSON library available for Python, while also being the least resource-intensive (especially regarding memory). It achieves this by virtue of being a light wrapper around Rust's https://serde.rs

Other Python projects are beginning to recommend and use it too

See: https://github.com/ijl/orjson#performance

[codecov]

Codecov Report

Merging #6143 into master will decrease coverage by 0.76%.
The diff coverage is 54.54%.

Impacted Files	Coverage Δ
...ks_base/datadog_checks/base/utils/serialization.py	0.00% <0.00%> (ø)
twistlock/datadog_checks/twistlock/twistlock.py	74.12% <100.00%> (+0.11%)	⬆️
twistlock/tests/test_twistlock.py	98.24% <100.00%> (-1.76%)	⬇️
datadog_checks_dev/tests/test_docker.py	48.14% <0.00%> (-51.86%)	⬇️
datadog_checks_dev/tests/tooling/test_utils.py	73.40% <0.00%> (-26.60%)	⬇️
datadog_checks_dev/datadog_checks/dev/env.py	22.72% <0.00%> (-20.46%)	⬇️
datadog_checks_dev/datadog_checks/dev/docker.py	24.57% <0.00%> (-19.50%)	⬇️
...atadog_checks_dev/datadog_checks/dev/structures.py	68.85% <0.00%> (-18.04%)	⬇️
datadog_checks_dev/tests/test_conditions.py	82.25% <0.00%> (-17.75%)	⬇️
datadog_checks_dev/datadog_checks/dev/_env.py	54.79% <0.00%> (-9.59%)	⬇️
... and 366 more

[florimondmanca]

Great idea, I've heard good things about orjson too, and I agree adoption is growing (eg see FastAPI).

Quick idea for converting this to a base import for frictionless usage in other integrations…

[florimondmanca]

Nice stuff!