fix(redfish): gate hgx diagnostic plan-b by debug toggle

bible-local/03-api.md

@@ -58,6 +58,7 @@ Responses:
 
 Optional request field:
 - `power_on_if_host_off`: when `true`, Redfish collection may power on the host before collection if preflight found it powered off
+- `debug_payloads`: when `true`, collector keeps extra diagnostic payloads and enables extended plan-B retries for slow HGX component inventory branches (`Assembly`, `Accelerators`, `Drives`, `NetworkAdapters`, `PCIeDevices`)
 
 ### `POST /api/collect/probe`
 

bible-local/05-collectors.md

@@ -27,6 +27,7 @@ Request fields passed from the server:
 - credential field (`password` or token)
 - `tls_mode`
 - optional `power_on_if_host_off`
+- optional `debug_payloads` for extended diagnostics
 
 ### Core rule
 
@@ -57,6 +58,17 @@ closes `skipCh` → goroutine in `Collect()` → `cancelCollect()`.
 
 The skip button is visible during `running` state and hidden once the job reaches a terminal state.
 
+### Extended diagnostics toggle
+
+The live collect form exposes a user-facing checkbox for extended diagnostics.
+
+- default collection prioritizes inventory completeness and bounded runtime
+- when extended diagnostics is off, heavy HGX component-chassis critical plan-B retries
+  (`Assembly`, `Accelerators`, `Drives`, `NetworkAdapters`, `PCIeDevices`) are skipped
+- when extended diagnostics is on, those retries are allowed and extra debug payloads are collected
+
+This toggle is intended for operator-driven deep diagnostics on problematic hosts, not for the default path.
+
 ### Discovery model
 
 The collector does not rely on one fixed vendor tree.

bible-local/10-decisions.md

@@ -1120,3 +1120,20 @@ incomplete for UI and Reanimator consumers.
 - System firmware such as BIOS and iBMC versions survives xFusion file exports.
 - xFusion archives participate more reliably in canonical device/export flows without special UI
   cases.
+
+---
+
+## ADL-043 — Extended HGX diagnostic plan-B is opt-in from the live collect form
+
+**Date:** 2026-04-13
+**Context:** Some Supermicro HGX Redfish targets expose slow or hanging component-chassis inventory
+collections during critical plan-B, especially under `Chassis/HGX_*` for `Assembly`,
+`Accelerators`, `Drives`, `NetworkAdapters`, and `PCIeDevices`. Default collection should not
+block operators on deep diagnostic retries that are useful mainly for troubleshooting.
+**Decision:** Keep the normal snapshot/replay path unchanged, but gate those heavy HGX
+component-chassis critical plan-B retries behind the existing live-collect `debug_payloads` flag,
+presented in the UI as "Сбор расширенных данных для диагностики".
+**Consequences:**
+- Default live collection skips those heavy diagnostic plan-B retries and reaches replay faster.
+- Operators can explicitly opt into the slower diagnostic path when they need deeper collection.
+- The same user-facing toggle continues to enable extra debug payload capture for troubleshooting.