Fix bytes.Reader issues

[ckbaker10]

Checklist

• make test-all (UNIX) passes. CI will also test this
• unit and/or integration tests are included (if applicable)
• documentation is changed or added (if applicable)

Description of change

file resource contents checks always reported object: *bytes.Reader on failure instead of the actual file content, making it completely useless for debugging.

Root cause: in ValidateGomegaValue, the func() (io.Reader, error) switch arm called f() and passed the raw io.Reader as foundValue to HavePatterns. The matcher scanned it fine, but once consumed, FailureResult had nothing left to read and fell through to fmt.Sprintf("object: %T", actual).

Fix: materialize the reader into a string immediately in the switch arm before handing it to the matcher. HavePatterns already handles string input, and FailureResult already formats strings correctly — so no changes needed there.

Updated three golden test fixtures that were asserting the broken "object: *strings.Reader" output.

[kgaughan]

I'm hoping I can get this merged in a subsequent release. It's borderline between a bugfix and a behavioural change, but it's definitely worth fixing, if only because I hit this issue when fixing other bugs.

In the meantime, would you mind merging from master?

[kgaughan]

@ckbaker10 Now that 0.4.10 is out and everything's been fixed up, I'd be very interested in getting this merged. The same goes for #1057. However, owing to the vintage of the two PRs, I need you to do a merge from master to pick up the updated integration test pipelines, &c. This way, I can kick them off to make sure everything's good.