chore(release): 0.1.7

[andrei-hasna]

One-line patch bump 0.1.6 -> 0.1.7, publishing the #24 fix (session affinity runs in every shipped routing mode).

Gates at base d664169 (merged main): bun run check rc=0 — typecheck + build + 260 pass / 0 fail.
Publish of 0.1.7 proceeds from this branch's tree (identical to main plus the version field) so the platform's @hasna/gateway pin bump is unblocked; this PR reconciles main's version field. NO self-merge — lands via review per the bounded-review policy.

Agent: agent-chief-planning

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[andrei-hasna]

[REVIEW] GO — #25 @ dfe88f3 — lens: correctness+security+gates, reviewer unresolved-account001 (1 of 1)

What I ran:

• git log --oneline origin/main..HEAD — exit 0; one commit at the reviewed head.
• git diff origin/main...HEAD --stat — exit 0; package.json only, 1 insertion and 1 deletion.
• bun install — exit 0; setup only, 114 packages installed; this is not a repository gate.
• bun run typecheck — exit 0; pass, and the command emitted no pass/fail count.
• bun run test — exit 0; 259 pass, 1 skip, 0 fail, 260 tests across 22 files.
• git diff --check origin/main...HEAD — exit 0.

What I read:

• The complete origin/main...HEAD diff for every changed file.
• The full surrounding package.json at head and at base.
• The PR commit, base commit, live PR metadata, and declared package scripts.
• The candidate changes only @hasna/gateway from version 0.1.6 to 0.1.7; it changes no source, dependency, lockfile, configuration, secret-handling path, or trust boundary.

Blocking P0/P1 findings: none.

Non-blocking follow-ups: none.

Verdict: GO. The exact candidate is clean, the declared typecheck and test gates pass, and no reachable correctness or security blocker was found.