[Bug] homepage should not be downloaded when checkver script is specified

[kazzarin]

Bug Report

Current Behavior

When checkver.script is specified in a manifest and checkver.url is not, checkver defaults to requesting the homepage url before running the script. If that request errors, the script is never run and checkver fails.

Expected Behavior

When checkver.script is specified, homepage and checkver.url should be ignored and the script should run as expected.

Additional context/output

I changed the homepage for a manifest in my own bucket, which produced the following error:

> .\bin\checkver.ps1 oracle-graalvm
oracle-graalvm: The operation has timed out.
URL https://www.oracle.com/java/graalvm/ is not valid

Reverting the change fixed it. The same behaviour was observed in my bucket's excavator, so I'm sure it's not a problem with my local machine. The wiki docs on the script property don't mention if it can be used with checkver.url or homepage, so I assume based on the manifests in the given examples that checkver.script is intended to be used exclusively when specified.

Unrelated, the site seems to timeout because of the link in scoop's user agent.

Possible Solution

System details

Windows version: 11

OS architecture: 64bit

PowerShell version: 7.3.9

Additional software: Git

Scoop Configuration

{
    "aria2-enabled":  false,
    "aria2-warning-enabled":  false,
    "cat_style":  "plain",
    "debug":  true,
    "gh_token":  "redacted",
    "last_update":  "2023-10-27T22:05:56.4511315-04:00",
    "scoop_branch":  "develop",
    "scoop_repo":  "https://github.com/ScoopInstaller/Scoop",
    "show_update_log":  false,
    "use_external_7zip":  true
}