chore(release): v2.92.1

Hotfix - Foundry quota preflight (last open Section 14 Priority D item).

Phase-FOUNDRY-QUOTA-PREFLIGHT shipped via 5-slice chain in 35.6 minutes.
Opt-in via PFORGE_FOUNDRY_QUOTA_PREFLIGHT=warn|block. Reads AOAI deployment
quota via Azure Cognitive Services control-plane REST API. Fail-open
invariant - control-plane outages NEVER block plan execution.

Cost: 0.05 USD declared / 0.00 USD wall (gh-copilot subscription).
costForLeg() byte-identical to v2.92.0; priceSlice() signature unchanged.
34 new tests pass; full test suite unaffected.

Author: srnichols

CHANGELOG.md

@@ -9,6 +9,10 @@ Format follows [Keep a Changelog](https://keepachangelog.com/).
 
 ---
 
+## [2.92.1] — 2026-05-08 — Hotfix: Foundry quota preflight
+
+> **One-liner**: Hotfix release adding opt-in AOAI quota preflight for plans running on the `microsoft-foundry` provider. Reads deployment TPM capacity via the Azure Cognitive Services control-plane API, compares against slice token estimates, and emits warnings (or optionally blocks execution) before a slice runs. Fail-open invariant: control-plane outages NEVER block execution. Closes the last open `docs/research/enterprise-fleet-readiness.md` §14 Priority-D item.
+
 ### Phase-FOUNDRY-QUOTA-PREFLIGHT — Azure AI Foundry deployment quota preflight
 
 > **One-liner**: Adds a quota preflight step to `forge_run_plan` for Microsoft Foundry / BYO Azure OpenAI deployments — fetches TPM capacity from the Azure Cognitive Services control-plane REST API, caches results for 5 minutes, compares the slice token estimate against available headroom (safe ≥ 30 %, warning 10–30 %, critical < 10 %), and logs a structured `[foundry-quota]` annotation on every slice. Fail-open: any quota fetch error produces `status: "unknown"` and never blocks execution. Block mode (`PFORGE_FOUNDRY_QUOTA_PREFLIGHT=block`) halts execution on `critical` status.

VERSION

@@ -1 +1 @@
-2.93.0-dev
+2.92.1

pforge-mcp/package.json

@@ -1,6 +1,6 @@
 {
   "name": "plan-forge-mcp",
-  "version": "2.93.0-dev",
+  "version": "2.92.1",
   "description": "MCP server exposing Plan Forge operations as tools for AI agents",
   "main": "server.mjs",
   "type": "module",