fix(zod-openapi): reduce route type instantiations

[thekiwi]

Description

This PR reduces the cost of type-checking apps that use createRoute,.openapi() and RouteHandler by restructuring a few of the route-typing generics. There is no runtime change, the emitted JS is identical. The inferred route/handler/response/RPC types are also unchanged.

Relates to #1918.

Problem

The route-typing generics re-derive several expensive sub-expressions on every use:

• ExtractStatusCode<Status> (which expands ranges like '5XX') is re-computed in every response branch, dragging a fresh TypedResponse cascade through each comparison.
• the per-route input intersection (param/query/header/cookie/form/json) and the response-shape conditional are inlined verbatim at around 5 sites.
• the request-body schema is walked (...['content'][keyof ...]['schema']) ~6 times per input type.

On large RPC apps, this makes type-checking more expensive than it needs to be.

Solution

Bind each expensive sub-expression once (infer ... extends ...) and share the duplicated pieces, with no change to the resulting types:

• resolve ExtractStatusCode<Status> once per response branch and reuse the inferred Content (the bulk of the win)
• share the input intersection via the existing ComputeInput<R> and the response-shape conditional via a new HandlerResponse<R> across RouteHandler / RouteHook / openapi() / HandlerFromRoute / HookFromRoute
• bind the request/response schema once in InputTypeBase / InputTypeJson / InputTypeForm / ExtractContent
• drop the duplicate RoutingPath (identical to the existing ConvertPathType)

Results

While the majority of the benchmarking was done with a sizeable closed-source project, I also generated a 100-route synthetic RPC app (CRUD + auth/upload/search/status) and measured with tsc --extendedDiagnostics. Instantiations & Types are deterministic; check time / memory are medians of 5 runs.

Metric	TS 6.0.3 (before → after)	TS 7.0.1-rc (before → after)
Instantiations	1,137,695 → 450,121 (-60%)	1,294,335 → 605,109 (-53%)
Types	284,671 → 199,940 (-30%)	379,680 → 297,250 (-22%)
Check time	3.00s → 1.10s (-63%)	0.587s → 0.266s (-55%)
Memory	~449 → ~295 MB (-34%)	~196 → ~171 MB (-13%)

The benchmark file is available here: https://gist.github.com/thekiwi/d85b3a0e5ff467c8ffcf21b9e96a8732

Safety / behaviour-neutral

• Existing runtime + *.test-d.ts suites pass
• Added type-level tests pinning the inferred types on the touched paths: text/plain / application/*+json / no-content / default responses, the 5XX status range, form/header/cookie input, and the typed-vs-raw-Response handler return
• Since the change is type-only, the runtime suite can't catch a regression, so I verified the new type tests actually fail on a deliberate break (I manually mutation-tested the restructured branches, all mutations were caught).

The author should do the following, if applicable

• Add tests
• Run tests
• yarn changeset at the top of this repo and push the changeset
• Follow the contribution guide

[changeset-bot]

🦋 Changeset detected

Latest commit: d9561a8

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@hono/zod-openapi	Patch

Not sure what this means? Click here to learn what changesets are.

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.05%. Comparing base (a401fee) to head (d9561a8).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1988   +/-   ##
=======================================
  Coverage   92.05%   92.05%           
=======================================
  Files         115      115           
  Lines        4065     4065           
  Branches     1059     1059           
=======================================
  Hits         3742     3742           
  Misses        287      287           
  Partials       36       36           

Flag	Coverage Δ
zod-openapi	94.90% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.