Fusion x BigQuery issue: incompatible nanosecond-grain timestamps written to elementary.dbt_run_results blocking edr send-report

[niccoloalexander]

Description

The on-run-end hook hangs for 8+ minutes when using dbt Fusion (v2.0.0-preview.149) with Elementary v0.23.0 on a large project (~2,600 models). The same project completes the on-run-end in a reasonable time with dbt Core 1.11.6.

Environment

Elementary version: 0.23.1
dbt Fusion version: 2.0.0-preview.154
dbt Core version (where it works): 1.11.6
Adapter: BigQuery 1.11.0

Sample Error

�[31;20m2026-04-08 23:24:18 — ERROR — Could not generate the report - Error: Failed to run dbt command. {"info": {"name": "RunningOperationCaughtError", "code": "Q001", "msg": "Encountered an error while running operation: Database Error\n Invalid timestamp: '2026-04-08T09:08:47.757610181Z'", "level": "error", "invocationId": "134aa3a4-375b-41b1-b666-cc15868994b0", "pid": 451, "thread": "MainThread", "ts": "2026-04-08T23:24:17.783811Z"}, "data": {"exc": "Database Error\n Invalid timestamp: '2026-04-08T09:08:47.757610181Z'"}}

Root Cause Analysis

Summary

The failure is caused by malformed timestamp precision in Elementary artifact tables when runs are executed with dbt Fusion.

What happened

edr send-report runs elementary_cli.get_test_results, which reads from elementary.dbt_run_results and applies timestamp filters/casts (notably on execute_completed_at).
Some rows written by dbt Fusion contain timestamp strings with 9 fractional digits (nanoseconds), e.g.:
2026-04-08T09:08:47.757610181Z
BigQuery TIMESTAMP parsing in this path rejects those values, producing:
Invalid timestamp ... while executing the filter on column 'execute_completed_at'

Scope / pattern

Same project path with dbt Core 1.11.6 did not produce these malformed values.
dbt Fusion 2.0.0 runs produce the malformed values
Using the forked elementary dbt package referenced in the Fix PR below does not produce the malformed values

Suggested Fix

Shared a PR#977 for review