perf(tracing): skip span-start upsert by default (end-only ingest) #1644

Job	Run time
find-tutorials	6s
test-10_async/00_base/080_batch_events	3m 41s
test-10_async/00_base/010_multiturn	2m 53s
test-10_async/00_base/030_tracing	2m 7s
test-10_async/10_temporal/010_agent_chat	3m 5s
test-00_sync/000_hello_acp	2m 27s
test-10_async/00_base/000_hello_acp	2m 43s
test-00_sync/030_langgraph	3m 27s
test-10_async/00_base/110_pydantic_ai	2m 19s
test-10_async/00_base/100_langgraph	3m 4s
test-10_async/10_temporal/020_state_machine	3m 30s
test-00_sync/040_pydantic_ai	2m 17s
test-10_async/00_base/020_streaming	3m 1s
test-10_async/10_temporal/070_open_ai_agents_sdk_tools	2m 49s
test-10_async/00_base/040_other_sdks	4m 22s
test-00_sync/020_streaming	3m 17s
test-10_async/00_base/120_openai_agents_local_sandbox	4m 24s
test-10_async/10_temporal/050_agent_chat_guardrails	2m 9s
test-10_async/10_temporal/060_open_ai_agents_sdk_hello_world	2m 29s
test-10_async/10_temporal/110_pydantic_ai	2m 35s
test-10_async/10_temporal/000_hello_acp	2m 9s
test-10_async/10_temporal/090_claude_agents_sdk_mvp	2m 10s
test-10_async/10_temporal/120_openai_agents_local_sandbox	2m 30s
test-10_async/10_temporal/130_langgraph	2m 23s
test-00_sync/010_multiturn	3m 10s
test-10_async/10_temporal/030_custom_activities	2m 15s
test-00_sync/050_openai_agents_local_sandbox	2m 24s
test-10_async/10_temporal/080_open_ai_agents_sdk_human_in_the_loop	2m 19s
Test Summary	4s
	1h 16m 9s

Provide feedback