build(profile): compile non-kernel workspace crates at dev opt-level 0

[devin-ai-integration]

Summary

• Compiles every non-kernel tracedecay-* workspace crate at dev opt-level = 0 (40 crates), mirroring the existing tracedecay-contracts / tracedecay-daemon-protocol precedent.
• Adds [profile.perf.package.*] opt-level = 1 mirrors so perf (which inherits dev overrides) keeps every crate at its previous effective level.
• Cold cargo build -p tracedecay-cli on 8 cores: 461s → 345s wall (−25%), workspace CPU 2156s → 1178s (−45%). One timed run each.

Motivation

A --timings critical-path audit (walking unblocked_units/unblocked_rmeta_units backward from the last unit) shows the wall clock is a serial chain of workspace crates whose cost is ~85% LLVM codegen at dev opt-level = 1:

schemars proc-macros (16s) → domain 107s → contracts 87s → rusqlite-runtime 103s
→ runtime-core 43s → daemon-identity/control → daemon-service 61s → mcp 45s
→ tracedecay 98s → tracedecay-cli link 18s

Frontend time is small (5-29s per crate); codegen is 33-96s. tracedecay (the root crate) is the slowest gating crate near the tail (98s) because it depends on all 64 workspace crates and can only start once every rmeta is ready; tracedecay-application (158s) is the slowest overall crate but finishes in parallel and does not gate.

Measured: dev opt-level = 0 on the non-kernel crates removes ~116s of wall time. The kernel crates that are perf-pinned (code-index, code-index-runtime, domain, graph-db, runtime-core, rusqlite-runtime, store, code-extraction, grafeo-*) keep dev opt-level = 1, so local cargo test still runs optimized engine code.

For comparison, dropping the kernels to opt-level = 0 too reaches 257s wall (−44%), at the cost of running all engine code unoptimized in local dev/test builds, left out of this PR.

Changes

• Cargo.toml: [profile.dev.package.*] opt-level = 0 for 40 non-kernel workspace crates plus [profile.perf.package.*] opt-level = 1 mirrors.

Test plan

• cargo build -p tracedecay-cli passes and produces a working binary (timed run above)
• cargo nextest run --workspace --no-fail-fast passes (CI)
• cargo clippy has no new warnings (no code changes)

Checklist

• CHANGELOG.md updated, N/A, build profile only
• No secrets, credentials, or .env files included
• Breaking changes documented, none; dev-only codegen quality, no API/behavior change

Link to Devin session: https://app.devin.ai/sessions/389efda907324a53bdd07ef9f4f29802
Open in Devin Desktop: https://app.devin.ai/desktop/session/389efda907324a53bdd07ef9f4f29802?variant=devin
Requested by: @ScriptedAlchemy

Link to Devin session: https://app.devin.ai/sessions/d1c43c1ea5264a0e9398d1b722b54447
Open in Devin Desktop: https://app.devin.ai/desktop/session/d1c43c1ea5264a0e9398d1b722b54447?variant=devin

---

[devin-ai-integration]

I'll fix CI failures and address comments from users with write access. I'll skip comments containing "(aside)".

• Disable automatic comment, CI, and merge conflict monitoring

[changeset-bot]

⚠️ No Changeset found

Latest commit: 49c1344

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no bugs or issues to report.