The existing ADF contract is inadequate to rigorously parse

Issue Summary

When parsing ADF, external developers need to choose between streaming or in-memory techniques. In memory techniques necessitate an upper bound for the size of the ADF to be specified in the API. Streaming techniques require more rigorous ordering of nodes to be specified in the API. For example, the `type` node should be streamed first to ensure subsequent nodes are fully interpretable.

See https://community.developer.atlassian.com/t/what-is-the-maximum-size-of-an-adf-document-i-e-should-we-use-a-streaming-parser/39079 for additional context.

Steps to Reproduce

Not applicable

Expected Results

Not applicable

Actual Results

Not applicable

Workaround

Not applicable