feat: Bitpacked Trio 16-bit layout [u16; 3] for O(1) graph traversal

[AdaWorldAPI]

The Structural Foundation

This defines the 16-bit address layout required for the Bitpacked Trio architecture:

Edge Triple: [Node_u16, Edge_u16, Node_u16] = 48 bits
With padding: 64 bits = perfect cache line alignment

Address Space (65,536 total)

Zone	Prefix	Count	Purpose
OPS	0x00-0x0F	4,096	Instruction set, prefix executes
	0x0C-0x0E	768	USER RESERVED
	0x0F:XX	256	Learned/custom thinking styles
FLUID	0x10-0x3F	12,288	Registers, cache, crystal, stack
EDGES	0x40-0x7F	16,384	Relationship envelope [Node, EDGE, Node]
NODES	0x80-0xFF	32,768	Dn tree topology [NODE, Edge, NODE]

Why NOT u24

• No hardware u24 → pay for u32 or burn cycles masking
• Breaks [u16; 3] alignment → kills AVX-512 throughput
• Scale via Shard_ID: [Shard_u16 | Local_u16] not wider pointers

Key Features

• EdgeTriple struct: 64-bit aligned, pack/unpack for AVX-512
• Dn tree addressing: Tree.Branch.Twig.Leaf with 2^4 branching
• Hamming distance = topological distance (XOR + popcount)
• Sublanguage support: SQL/CQL, Cypher/GQL, NARS/ACT-R slot subdivision
• User reserved: 0x0C-0x0E for custom extensions
• Learned styles: 256 slots at 0x0F for custom cognition

Removes

• Grammar Layer (NSM, Templates, Speech Acts) - moved to user space or edges
• Old SURFACE/FLUID/NODE naming → OPS/FLUID/EDGES/NODES

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.