fix: use reference outputs for example creation [closes LSE-2783]

[Zhen (yumuzi234)]

Description

Python and TypeScript example creation APIs now prefer reference-output terminology while retaining deprecated outputs aliases. Canonical JSON payloads use reference_outputs; multipart storage names remain unchanged. Merge and deploy the API compatibility PR before releasing this change.

Linear: https://linear.app/langchain/issue/LSE-2783/update-dataset-api-and-sdk-to-use-reference-outputs-instead-of-outputs
Depends on: https://github.com/langchain-ai/langchainplus/pull/34107 · UI: https://github.com/langchain-ai/langchainplus/pull/34106

Test Plan

• Verify canonical, legacy, and conflicting values in Python and TypeScript client tests.
• Run both SDK unit suites.

Made by Open SWE

[github-actions]

JS perf benchmark

Lower is better. Noisy on shared runners — treat as a signal, not a gate.

Base64-heavy payload

Single large base64 string per message — the shape the worker-offload path is optimized for.
Payload: 2511.2 KB in / 5.2 KB out, 100 runs.

metric	main	this PR	delta
Wall time (ms)	1590.67	1812.92	+14.0%
createRun total (ms)	94.13	100.43	+6.7%
createRun p50 (ms)	0.83	0.83	+0.8%
createRun p95 (ms)	1.11	1.86	+67.6%
createRun p99 (ms)	7.72	7.76	+0.6%
createRun max (ms)	7.72	7.76	+0.6%
updateRun total (ms)	31.49	29.71	-5.7%
updateRun p95 (ms)	0.79	0.55	-29.9%
loop lag total (ms)	826.50	978.42	+18.4%
loop lag p50 (ms)	0.11	0.11	-3.2%
loop lag p95 (ms)	4.14	4.98	+20.3%
loop lag p99 (ms)	18.99	11.84	-37.6%
loop lag max (ms)	77.99	74.10	-5.0%

Structural payload

Many small strings across a wide/nested object graph. Should bypass worker offload and use sync flush.
Payload: 1239.5 KB in / 13.3 KB out, 100 runs.

metric	main	this PR	delta
Wall time (ms)	2023.34	1815.18	-10.3%
createRun total (ms)	679.82	630.49	-7.3%
createRun p50 (ms)	6.02	5.71	-5.3%
createRun p95 (ms)	14.10	11.92	-15.5%
createRun p99 (ms)	21.94	12.28	-44.0%
createRun max (ms)	21.94	12.28	-44.0%
updateRun total (ms)	36.15	34.31	-5.1%
updateRun p95 (ms)	0.50	0.52	+4.3%
loop lag total (ms)	1587.03	1414.63	-10.9%
loop lag p50 (ms)	0.09	0.10	+4.2%
loop lag p95 (ms)	13.42	11.22	-16.4%
loop lag p99 (ms)	101.42	135.13	+33.2%
loop lag max (ms)	201.87	155.75	-22.8%