ai-memory: back the demo with real Context Graph packages

[antejavor]

Summary

code-examples/ai-memory.md/.sh/.ps1 illustrated Memgraph's semantic/episodic/procedural memory pitch with a hand-rolled Client/Preference/Interaction/Workflow/Step schema and the generic mcp-memgraph server — no connection to how Memgraph's own Context Graph project actually implements durable AI memory.

This rewires the demo to write and recall all three memory types through the real packages a live coding-assistant plugin uses:

• sessions-graph — semantic memory (Memory nodes)
• actions-graph — episodic memory (Session/Action nodes, FOLLOWED_BY sequencing)
• skills-graph — procedural memory (Skill nodes, USED_SKILL)

These three share a (:User)/(:Session) node as join key, so the demo's "interconnected recall" payoff is now a genuine one-shot Cypher traversal through that shared graph, not three lookups glued together in script. The "Wire It Into a Real Harness" section now points at the actual agent-context-graph plugin bootstrap flow (whose Memgraph defaults already match this demo's container) instead of a generic MCP config snippet.

This trades the previous "no Python" positioning for Python 3.10-3.13 (to pip install the three packages into a throwaway venv) — a deliberate scope call to make the example true to the real product, discussed and confirmed before implementation.

• New ai-memory.py holds the seed/recall logic.
• ai-memory.sh/.ps1 now provision that venv alongside the same single Memgraph container (no more separate mcp-memgraph/mgconsole containers — schema-info/ontology inspection reuses the mgconsole already bundled in memgraph-mage).
• ai-memory.md rewritten to match.

Test plan

• Ran ./ai-memory.sh end-to-end against a live Docker Memgraph container: network/container up, venv created, sessions-graph/actions-graph/skills-graph installed from PyPI, seed + all four recall queries (semantic/episodic/procedural/interconnected) produced correct output, SHOW SCHEMA INFO succeeded, wrap-up message printed.
• Ran ./ai-memory.sh clean: container, network, and venv removed.
• Ran the PowerShell counterpart under a real pwsh 7.4.6 binary (not just syntax-parsed): full run flow reproduced identical output, clean flow removed everything.
• [System.Management.Automation.Language.Parser]::ParseFile confirms ai-memory.ps1 has no syntax errors.
• Found and fixed a real ordering bug during testing: sessions-graph's save_memory(session_id=...) provenance write races actions-graph's create_session unique constraint on Session if session creation doesn't happen first — ai-memory.py now creates sessions before anything else references them.