feat(ecommerce): cutover to MAF runtime

[Cataldir]

Problem statement

Migrate the ecommerce-* bounded context (5 services: cart-intelligence, catalog-search, checkout-support, order-status, product-detail-enrichment) from the legacy BaseRetailAgent runtime internals to the MAF-backed runtime introduced in #981. This is the largest single bounded context in R1 (5 services in one cutover).

This is a hard sunset: legacy runtime imports are deleted in the same PR. There is no coexistence flag.

Required change

For each of ecommerce-cart-intelligence, ecommerce-catalog-search, ecommerce-checkout-support, ecommerce-order-status, ecommerce-product-detail-enrichment:

• Replace usage of lib/src/holiday_peak_lib/agents/ internals with the public surface backed by the MAF runtime from feat(lib): MAF runtime adapter behind the existing BaseRetailAgent surface #981.
• Update each main.py and the agent definition to use the MAF-backed seam.
• MCP tool registrations via FastAPIMCPServer remain unchanged in signature; only the underlying runtime changes.
• REST endpoints remain unchanged in shape; migration is internal.
• Per-service Foundry config (FOUNDRY_AGENT_ID_FAST, MODEL_DEPLOYMENT_NAME_FAST, FOUNDRY_AGENT_ID_RICH, MODEL_DEPLOYMENT_NAME_RICH, FOUNDRY_STREAM, PROJECT_ENDPOINT) wired via Helm values + Key Vault refs.
• Three-tier memory bound per ADR-032 (PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978).
• MCP-only A2A enforced (ADR-030, PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978); hop counter wired.
• Delete pre-MAF imports across all five services in the same PR.

Affected components

• apps/ecommerce-cart-intelligence
• apps/ecommerce-catalog-search
• apps/ecommerce-checkout-support
• apps/ecommerce-order-status
• apps/ecommerce-product-detail-enrichment

Acceptance criteria (per spec — Acceptance gate per bounded context)

• Contract tests in lib/tests pass.
• All apps/ecommerce-* unit and integration tests pass.
• Continuous eval baseline (per ADR-028 final schema, PR Add continuous agent evaluation engine #974) within maxDeltaPercent for each agent in this context (advisory until PR Add continuous agent evaluation engine #974 merges).
• AGC canary 5 → 25 → 50 → 100 succeeds in dev, then in staging (per ADR-029, PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978).
• No A2A HTTP detected by scripts/ci/lint_no_a2a_http.py (per ADR-030, PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978).
• Span attribute contract validated by App Insights workbook query (per ADR-031, PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978).
• No pre-MAF imports remain in apps/ecommerce-cart-intelligence, apps/ecommerce-catalog-search, apps/ecommerce-checkout-support, apps/ecommerce-order-status, apps/ecommerce-product-detail-enrichment.
• Per-service Foundry config wired via Helm values + Key Vault refs for all five services.
• 24-hour steady-state hold satisfied before the next context's PR opens.

Risks and dependencies

Risk	Mitigation
ecommerce is the largest bounded context — 5 services in one cutover.	Pre-flight grep + dedicated contract tests cover the cross-service browse → cart → checkout → order journey. tests/e2e/ end-to-end retail journey runs in CI before the canary opens.
Checkout-support regressions risk transactional flow failures (high blast radius).	Canary ladder (5→25→50→100) gives early detection; contract tests cover checkout edge cases; outside tolerance halts the canary immediately.
Catalog-search relevance regresses on cutover.	Eval baselines cover relevance metrics; outside tolerance halts the canary.
MCP A2A introduces extra latency on the browse → cart → checkout chain.	Measured pre/post per service; tolerance P95 within 10 %.
Reach-throughs into BaseRetailAgent internals from any of the 5 services.	Pre-flight grep blocks cutover until refactored.
Foundry agent ID provisioning fails for any of the 5 services.	Provisioning runs ahead of this PR; failure halts this context.

Blocked by:

• feat(lib): MAF runtime adapter behind the existing BaseRetailAgent surface #981 (lib MAF runtime seam).
• feat(logistics): cutover to MAF runtime #987 (logistics-* cutover — preceding context per the R1 cutover order).

Blocks: the next context (product-management-*) per the R1 sequencing rule.

Evidence links

• apps/ecommerce-cart-intelligence
• apps/ecommerce-catalog-search
• apps/ecommerce-checkout-support
• apps/ecommerce-order-status
• apps/ecommerce-product-detail-enrichment
• apps/ecommerce-cart-intelligence/src
• apps/ecommerce-catalog-search/src
• apps/ecommerce-checkout-support/src
• apps/ecommerce-order-status/src
• apps/ecommerce-product-detail-enrichment/src
• tests/e2e
• lib/src/holiday_peak_lib/agents/base_agent.py

ADR impact

• ADR-007 (memory tiers) — refined by ADR-032 (PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978).
• ADR-009 (ACP catalog search) — preserved; internal migration only.
• ADR-014 (ACP frontend) — preserved; REST endpoint surface unchanged.
• ADR-024 → ADR-030 (PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978): MCP-only A2A enforced.
• ADR-029 (PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978): AGC canary gates the cutover.
• ADR-031 (PR docs(adr): ADR-029..033 - architecture contracts (canary, MCP A2A, OTEL, memory, UI on SWA) #978): span attribute contract validated.

Branch convention

feature/<issue-id>-ecommerce-maf-cutover (per ADR-018).

BPMN process

%%{init: {'theme':'base', 'themeVariables': {
  'primaryColor':'#FFB3BA',
  'primaryTextColor':'#000',
  'primaryBorderColor':'#FF8B94',
  'lineColor':'#BAE1FF',
  'secondaryColor':'#BAE1FF',
  'tertiaryColor':'#FFFFFF'
}}}%%
flowchart LR
  A[Analyze Current Code] --> B[Design Change]
  B --> C[Implement on Issue Branch]
  C --> D[Open PR]
  D --> E[Validation and Fixes]
  E --> F[Merge to Main]
  F --> G[Monitor Workflows]
  G --> H[Close Issue and Cleanup]

Loading