[Chien Yuan Chang (chienyuanchang)]

Summary

• Preserve sanitized service messages and nested error codes in terminal output and per-input reports for HTTP, polling, and inline failures.
• Explain ModelNotSupported with the reported model name and commands to inspect analyzer support and resource model mappings; warn that defaults changes are resource-wide.
• Show accurate success/failure/skip counts and identify failed input ordinals. Mention retained files only when successful results were actually saved.
• Share bounded diagnostic redaction, including reflected credentials, and explicitly mark missing or truncated details. Keep cu-cli/analyze-report/v1 unchanged.

Stacking

Stacked on #10784; the base is ai-cu/terminal-markdown-rendering so this PR contains only the diagnostic fix. Retarget to main after the parent merges.

Validation

• Full extension and nested REST-client go test -race ./... passed.
• Builds, vet, modernization checks, strict lint, scoped spelling, and whitespace checks passed.
• Regression coverage includes nested and unknown errors, missing details, all analysis output modes and reports, mixed batches, long outer diagnostics, and reflected API-key/Bearer credentials.
• Offline fixtures/fake transports only; no live Azure analysis or resource mapping changes.

Analyzer-specific doctor readiness checks remain a separate follow-up.

[azure-pipelines]

Azure Pipelines:
Successfully started running 1 pipeline(s).
26 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.