[Python] Optionally write chunk content to Qdrant payload

[pangwangshu]

Adds an optional content_payload_key field to QdrantWriteConfig. When set, the converter stores EmbeddableItem.content_string under that key in the point payload, next to the metadata, so points retrieved from a collection written by Beam carry the chunk text. BigQueryVectorWriterConfig and the Milvus writer already persist the content; the Qdrant writer stored only item.metadata.

• The default is None, which writes only the metadata as before, so there is no behavior change for existing users.
• A key that collides with a metadata key raises ValueError instead of overwriting it.
• An empty key is rejected in __post_init__.
• The item's metadata dict is copied, not mutated.
• As with the BigQuery writer, an item with no text or image URI raises ValueError from content_string when the key is set.

The two design questions in #40476 are still open: whether leaving the content out was deliberate, and whether an opt-in key is the right shape compared with always writing to a fixed "content" key. I am happy to change the name or the default based on review.

Testing

• qdrant_test.py: 5 new converter tests (default unchanged, key writes content, no metadata, collision, empty key). All 43 tests in the file pass locally.
• qdrant_it_test.py: new test_write_with_content_payload_key writes through the transform to a testcontainers Qdrant instance and asserts the payload read back is the metadata plus the content. All 7 tests in the file pass locally on DirectRunner.
• CHANGES.md updated.

AI assistance disclosure

Per the ASF generative tooling guidance: this change was written with the help of Claude Code (Claude Opus 5.5). I have reviewed the code and tests and ran them locally, and to my knowledge the change contains no third-party copyrighted material.

Generated-by: Claude Code (Claude Opus 5.5)

Fixes #40476

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 56.23%. Comparing base (76f6e6d) to head (30227d4).
⚠️ Report is 111 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff              @@
##             master   #40502      +/-   ##
============================================
+ Coverage     56.16%   56.23%   +0.06%     
  Complexity     2288     2288              
============================================
  Files          1124     1124              
  Lines        178336   178367      +31     
  Branches       1489     1489              
============================================
+ Hits         100170   100304     +134     
+ Misses        75645    75542     -103     
  Partials       2521     2521              

Flag	Coverage Δ
python	79.68% <100.00%> (+0.13%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Assigning reviewers:

R: @jrmccluskey for label python.

Note: If you would like to opt out of this review, comment assign to next reviewer.

Available commands:

• stop reviewer notifications - opt out of the automated review tooling
• remind me after tests pass - tag the comment author after tests pass
• waiting on author - shift the attention set back to the author (any comment or push by the author will return the attention set to the reviewers)

The PR bot will only process comments in the main thread (not review comments).

[jrmccluskey]

Broadly LGTM from a Python perspective, looping in a second set of eyes that is more familiar with RAG to hopefully answer some of your outstanding questions on the initial implementation