feat: partial success in trace ingestion

[shuiyisong]

I hereby agree to the terms of the GreptimeDB CLA.

Refer to a related PR or issue link (optional)

close #7875

What's changed and what's your intention?

This PR changes OTLP trace ingestion to preserve valid spans when part of a request fails, instead of failing the entire batch.

• add OTLP trace partial success responses with rejected span counts and summarized error messages
• ingest traces by OTLP resource/scope group, then subchunk each group for bounded writes
• retry deterministic chunk failures span-by-span to salvage valid spans
• discard ambiguous chunk failures without retrying and count those spans as rejected
• split trace row building so main trace rows and auxiliary trace tables are handled independently
• derive auxiliary trace tables only from spans whose main-table writes succeeded

PR Checklist

Please convert it to a draft if some of the following conditions are not met.

• I have written the necessary rustdoc comments.
• I have added the necessary unit tests and integration tests.
• This PR requires documentation updates.
• API changes are backward compatible.
• Schema or data changes are backward compatible.

[fengys1996]

@codex review

[gemini-code-assist]

Code Review

This pull request implements a chunked ingestion strategy for OTLP trace spans, featuring a fallback mechanism that retries individual spans when deterministic chunk failures occur. It introduces a TraceIngestOutcome struct to support OTLP partial success responses, allowing the server to report accepted and rejected spans to the client. The parsing logic has been refactored to preserve resource and scope groupings, and the system now updates auxiliary tables for services and operations based on successfully ingested spans. Review feedback focuses on refining the error classification logic to avoid wasteful retries on metadata errors, ensuring transient errors are propagated rather than swallowed to allow client-side retries, and optimizing the parsing process by removing redundant span count calculations.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6a05653073

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[killme2008]

LGTM

Recommend adding more tests that simulate span or chunk write failures.