[Linux][Docs] July 14 release note contradicts managed vLLM storage behavior

## Description

The published July 14 release note says verified managed vLLM storage shortages are advisory during express and other non-interactive setup. Current managed vLLM documentation says those paths stop after a verified insufficient-capacity warning, so the public documentation is internally contradictory.

Platform scope: Reproduced on Linux only; other platforms not tested

Regression: Unknown — earlier published documentation versions not tested

OpenShell issue: No

## Environment

```text
Device: Documentation validation host
OS: Ubuntu 24.04
Architecture: x86_64
Node.js: N/A — published documentation validation only
npm: N/A — published documentation validation only
Docker: N/A — published documentation validation only
OpenShell CLI: N/A — published documentation validation only
NemoClaw: v0.0.119 documentation
OpenClaw: N/A — published documentation validation only
```

## Steps to Reproduce

1. Open https://docs.nvidia.com/nemoclaw/user-guide/openclaw/release-notes/2026/7/14.
2. Read the managed vLLM storage preflight entry.
3. Open https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/local-inference/set-up-vllm.
4. Read the storage preflight behavior for express installation and other non-interactive setup.
5. Compare the two statements.

## Expected Result

The historical release note accurately states that express and other non-interactive setup stop after a verified insufficient-capacity warning unless the documented override is used.

## Actual Result

The release note states that verified shortages are advisory during express and other non-interactive setup, while the current setup page states that these paths stop after the verified warning.

## Logs

```text
Release note excerpt: Managed vLLM storage preflight treats verified shortages as advisory during express and other non-interactive setup while the checks mature.

Set Up vLLM excerpt: Express installation and other non-interactive setup stop after a verified warning that Docker image storage or model-cache storage is insufficient, before the guarded download starts.
```

Compare the managed vLLM storage preflight entry in the July 14 release notes with the storage behavior in the Set Up vLLM page, using the two linked documentation URLs as the starting points. Update the release note so it matches the current stop behavior and documented override, then verify that both pages describe the same behavior.

Written by the indexing model from the issue text.