fix: address Bugbot review on #49 (output-json early-failure JSON + rm partial-teardown msg)

[aptracebloc]

Addresses the two Bugbot findings on the develop→main sync (#49). Per our sync-PR convention, the fixes land on develop here; #49 (head = develop) picks them up automatically.

Fixes

1. JSON missing on early failures (dataset.go, feat(#29): --output-json for dataset push (PR-b) #38 code) — --output-json only emitted JSON at the dry-run + post-submit points; the ~18 earlier exitError returns left stdout empty, breaking … --output-json | jq. runDatasetPush now uses a named return + jsonEmitted flag + a defer that emits a {"status":"error",…,"exit_code":N} object on any early-failure path. Adds Error/ExitCode to the result shape + writePushErrorJSON; unit-tested.
2. Rm drops table before file delete (teardown.go, feat(#30): dataset rm — in-cluster teardown (table + PVC) [DRAFT] #40 code) — on a partial teardown (table dropped, file removal failed), the error now names the partial state and tells the user to re-run. Both ops are idempotent (DROP TABLE IF EXISTS / rm -rf), so a re-run completes the cleanup.

Verification

• make ci green; new TestRunDatasetPush_OutputJSONEarlyFailureEmitsJSON.
• Live: dataset push … --output-json with a bad table now prints a JSON error on stdout (was empty) and exits 2.

🤖 Generated with Claude Code

---

Note

Low Risk
CLI-only output and error-message changes; no auth or data-path logic changes beyond surfacing partial teardown state.

Overview
dataset push --output-json now always writes a JSON object to stdout, including failures before dry-run or post-submit (validation, discovery, staging, auth, etc.). runDatasetPush uses a named error return, a jsonEmitted flag, and a defer that calls writePushErrorJSON with status: "error", error, and exit_code. Success paths set jsonEmitted after existing emits. pushJSONResult gains optional error and exit_code fields and documents status error.

dataset rm treats partial teardown explicitly: if Teardown dropped the MySQL table but PVC file removal failed, the CLI returns exit 7 with a message that the table is gone and the user should re-run tracebloc dataset rm to finish file cleanup.

Adds TestRunDatasetPush_OutputJSONEarlyFailureEmitsJSON (invalid table → exit 2 + JSON on stdout).

^{Reviewed by Cursor Bugbot for commit b873bd0. Bugbot is set up for automated code reviews on this repo. Configure here.}

[LukasWodka]

👋 Heads-up — Code review queue is at 16 / 8

Above the WIP limit. The team convention is to review existing PRs before opening new work.

Open PRs currently in Code review (oldest first):

• averaging-service#102 — chore(kanban): declare staging as prod-truth branch via .kanban.yml · author: @LukasWodka · no reviewer assigned
• averaging-service#98 — ci: add fr-gate caller workflow · author: @LukasWodka · no reviewer assigned
• cli#49 — Sync develop → main (prepare next release) · author: @aptracebloc · no reviewer assigned
• client#199 — Release v1.4.5: promote develop → main · author: @saadqbal · no reviewer assigned
• client-runtime#79 — docs(#59): feasibility — delete pod_tokens (and the reconciler)? · author: @saadqbal · no reviewer assigned
• data-ingestors#146 — test(schema): lock category enum to the engine's TaskCategory (anti-drift) · author: @LukasWodka · reviewer: @saadqbal
• data-ingestors#148 — Develop to prod · author: @divyasinghds · reviewer: @saadqbal
• design-system#19 — fix: un-track coverage/ and node_modules/ from git · author: @LukasWodka · no reviewer assigned
• design-system#22 — ci: add Vitest test workflow · author: @LukasWodka · reviewer: @aptracebloc
• design-system#23 — ci: add fr-gate caller workflow · author: @LukasWodka · no reviewer assigned

Pull from review before opening new work. (This is a nudge from the kanban WIP check, not a block.)