Split full lint from local format

[karakanb]

Speeds up local formatting by removing duplicate formatter work and deferring expensive linters. Keeps the complete lint suite in CI.

[greptile-apps]

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
Makefile:120-124
**`wait` with no arguments always returns exit code 0**, so `lint-full` will report success even when either `golangci-lint` invocation finds real issues. Per POSIX and bash specifications, `wait` with no jobspec waits for all children and unconditionally returns 0. The original CI ran the two linters as sequential steps with direct exit-code propagation — the parallel refactor breaks that guarantee. The fix is to capture each PID and wait for them individually.

```suggestion
lint-full:
	@echo "$(OK_COLOR)>> [golangci-lint] running full lint suite$(NO_COLOR)"; \
	golangci-lint run --timeout 10m60s --build-tags="no_duckdb_arrow" ./... & PID1=$$!; \
	(cd semantic-engine && golangci-lint run --timeout 10m60s ./...) & PID2=$$!; \
	wait $$PID1; R1=$$?; \
	wait $$PID2; R2=$$?; \
	exit $$((R1 + R2))
```

_{Reviews (1): Last reviewed commit: "split full lint from local format" | Re-trigger Greptile}

[greptile-apps]

wait with no arguments always returns exit code 0, so lint-full will report success even when either golangci-lint invocation finds real issues. Per POSIX and bash specifications, wait with no jobspec waits for all children and unconditionally returns 0. The original CI ran the two linters as sequential steps with direct exit-code propagation — the parallel refactor breaks that guarantee. The fix is to capture each PID and wait for them individually.

Suggested change

	lint-full:
	@echo "$(OK_COLOR)>> [golangci-lint] running full lint suite$(NO_COLOR)" & \
	golangci-lint run --timeout 10m60s --build-tags="no_duckdb_arrow" ./... & \
	cd semantic-engine && golangci-lint run --timeout 10m60s ./... & \
	wait
	lint-full:
	@echo "$(OK_COLOR)>> [golangci-lint] running full lint suite$(NO_COLOR)"; \
	golangci-lint run --timeout 10m60s --build-tags="no_duckdb_arrow" ./... & PID1=$$!; \
	(cd semantic-engine && golangci-lint run --timeout 10m60s ./...) & PID2=$$!; \
	wait $$PID1; R1=$$?; \
	wait $$PID2; R2=$$?; \
	exit $$((R1 + R2))

Prompt To Fix With AI

This is a comment left during a code review.
Path: Makefile
Line: 120-124

Comment:
**`wait` with no arguments always returns exit code 0**, so `lint-full` will report success even when either `golangci-lint` invocation finds real issues. Per POSIX and bash specifications, `wait` with no jobspec waits for all children and unconditionally returns 0. The original CI ran the two linters as sequential steps with direct exit-code propagation — the parallel refactor breaks that guarantee. The fix is to capture each PID and wait for them individually.

```suggestion
lint-full:
	@echo "$(OK_COLOR)>> [golangci-lint] running full lint suite$(NO_COLOR)"; \
	golangci-lint run --timeout 10m60s --build-tags="no_duckdb_arrow" ./... & PID1=$$!; \
	(cd semantic-engine && golangci-lint run --timeout 10m60s ./...) & PID2=$$!; \
	wait $$PID1; R1=$$?; \
	wait $$PID2; R2=$$?; \
	exit $$((R1 + R2))
```

How can I resolve this? If you propose a fix, please make it concise.

[greptile-apps]

_{Reviews (2): Last reviewed commit: "propagate full lint failures" | Re-trigger Greptile}