workflows: add missing translation directions

[idy]

What changed

• add reusable AST translation workflows for Japanese to Chinese, Korean to Chinese, and Chinese to French
• expose the three workflows from RuntimeProfile/default
• keep the existing translation model alias and translator voice alias contract

Why

H106 needs seven translation directions. Four already existed in Raids and had been exercised by existing deployments; keeping the remaining generic directions in Raids avoids duplicating AST workflow definitions in each product deployment. Product RuntimeProfiles remain responsible for resolving translator to the product-specific voice.

Validation

• scripts/validate_catalog.py with PyYAML 6.0.3
• unittest discover with PyYAML 6.0.3
• 1036 YAML files and the public default bootstrap closure validated
• all 10 catalog validator tests passed

[idy]

@codex

[github-actions]

🤖 OpenAI PR review

Reviewed commit: 09dfb8c813

Reviewed range: be5b2eca68..09dfb8c813

Review mode: full

Session key: repo-1308402561-pr-10-v2

Generation key: 2f0dc369e655c2d55d825c34308d31766559c5c208c83c7eab602058215ef37d

Diff chunks: 1

Model: gpt-5.6-terra

Reasoning effort: medium

Review time: 8s

Input tokens: 43,400

Cached input tokens: 31,645

Cache write tokens: 11,743

Cache hit ratio: 72.9%

Output tokens: 560

Reasoning tokens: 180

Total tokens: 43,960

Estimated Codex credits: 1.142

Credit rate per 1M tokens: 62.5 input / 6.25 cached / 375 output

Per-chunk and aggregation usage

Turn key	Time	Input	Cached	Cache hit	Cache write	Output	Reasoning	Total	Est. credits
chunk:1/1:633e34853aac0dbbd4c8874fe914e7debf86237ed15e431abb75a7a4fdccb6d9	5s	20,374	9,423	46.3%	10,945	375	170	20,749	0.884
aggregate:2f0dc369e655c2d55d825c34308d31766559c5c208c83c7eab602058215ef37d	3s	23,026	22,222	96.5%	798	185	10	23,211	0.259

No actionable findings.

No actionable findings.