Scripts do not wait long enough for results when running with --fail-if-not-fully-cacheable

[erichaagdev]

The availabilityWaitTimeoutSecs of the API is always configured to be 10 seconds with no way to override it. After this 10 seconds, if there is still no build scan data available, the scripts will proceed without the performance characteristics populated. In most cases this is fine because you can use the corresponding build scan quick links to investigate.

However, having the API call succeed and the corresponding build data to be available to the scripts is critical when running with --fail-if-no-fully-cacheable. I propose we increase availabilityWaitTimeoutSecs when running with --fail-if-not-fully-cacheable (e.g. 30s, 60s, 120s).

https://github.com/gradle/gradle-enterprise-build-validation-scripts/blob/b5e54b97633dd851bb59b135538c722a01c7c5db/components/fetch-build-scan-data-cmdline-tool/src/main/java/com/gradle/enterprise/api/client/GradleEnterpriseApiClient.java#L42

This is related to but does not replace #493. This will ensure we get a response in more cases while #493 will fail the build if no response is received.