Protect `env:start` command against connection issues

[desrosj]

[59231] moved the composer update command from the test:php script to env:start to prevent the command from running every time unit tests are run.

ticket:57189#comment:22 from @flixos90 mentions a situation where packagist.org was experiencing some down time and caused the env:start command to fail. The reason for this can be hard to decipher and it's easy to assume that you are experiencing a Docker issue.

Ideally the composer update command catches errors and displays a more informative notice to the user. "There was a failure running composer update." with some guidance to fix the issue.

See #57189.

[iamadisingh]

Currently, the composer update command is run as part of the env:start npm script in package.json

"env:start": "node ./tools/local-env/scripts/start.js && node ./tools/local-env/scripts/docker.js run -T --rm php composer update -W",

To provide a clearer, actionable error message, we could move the composer update logic into the JS startup script start.js. This would allow us to run Composer using Node's spawnSync, check the exit code, and print a custom message if it fails (something like "There was a failure running composer update. This is often caused by network issues or Packagist.org downtime. Please check your internet connection and try again.").

This PR improves the developer experience for npm run env:start by providing clear, actionable error messages if composer update fails.

Trac ticket: https://core.trac.wordpress.org/ticket/62807