impl(bigquery): remove QueryReference in favor of QueryCreationMetadata

[alvarowolfx]

Previously was assumed that QueryID and JobID were mutually exclusive (they're once, but now it isn't anymore). QueryReference was created with that in mind, so it doesn't make sense anymore and customers can access QueryID and JobID via QueryCreationMetadata now, giving access to the full RPC responses, without additional abstractions.

Towards #5844

[gemini-code-assist]

Code Review

This pull request refactors the BigQuery query API by removing the QueryReference enum and its associated query_reference() method. Instead, it introduces a metadata() method on the Query struct that returns a reference to QueryCreationMetadata, allowing direct access to fields like query_id and job_reference. All examples, tests, and internal usages have been updated to align with this simplified design. I have no additional feedback to provide as the changes are clean and idiomatic.

[codecov]

Codecov Report

❌ Patch coverage is 0% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 96.26%. Comparing base (af7c971) to head (6d158eb).

Files with missing lines	Patch %	Lines
src/bigquery/src/query/query_handle.rs	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6317      +/-   ##
==========================================
- Coverage   96.27%   96.26%   -0.01%     
==========================================
  Files         283      282       -1     
  Lines       73333    73278      -55     
==========================================
- Hits        70599    70541      -58     
- Misses       2734     2737       +3     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dbolduc]

comment: it is a tiny bit sad that the mutual exclusivity of these things is not encoded in the type. I think it's fine though.

[alvarowolfx]

yeah, unfortunately was an error from the service side to not declared it nullable. It's a hard call to us to add an override to change the type, even that I would like to.

[dbolduc]

optional nit:

Suggested change

	let query_id = &metadata.query_id;
	assert!(!query_id.is_empty(), "expected non-empty query_id");
	assert!(!metadata.query_id.is_empty(), "expected non-empty query_id");