chore(deps): bump opendal to 0.58.1

[Xuanwo]

I hereby agree to the terms of the CLA available at: https://docs.databend.com/dev/policies/cla/

Summary

Bump Apache OpenDAL from 0.54.1 to 0.58.1 (released 2026-07-31), and object_store_opendal to 0.58.0 (0.58.1 is yanked on crates.io).

OpenDAL 0.55–0.58 rework the service/layer composition model (Access → Service), switch schemes to strings, move metadata timestamps to jiff/opendal::raw::Timestamp, replace HttpClient/HttpFetch with HttpTransport, and return a finished Operator from Operator::new (no .finish()).

Fallout fixes

• Rewrite Databend custom layers (RuntimeLayer, ConcurrentLimitLayer, StorageMetricsLayer, metrics observe layer) for the new Layer / Service APIs
• allow_anonymous → skip_signature; full_capability → capability
• Scheme enum matching → string schemes / services::*_SCHEME
• Timestamp bridge via opendal_timestamp_to_chrono
• Align fastrace / fastrace-opentelemetry / opentelemetry so layers-fastrace builds with OpenDAL 0.58

Residual risk (blocked on ecosystem)

object_store_opendal 0.58 requires object_store 0.13, while current deltalake (git rev) and datafuse-extras/lance still pin object_store 0.12. That leaves:

• databend-common-storages-delta
• databend-common-storages-stage (Lance path via OpendalStore)

failing with OpendalStore: ObjectStore trait version mismatch until those crates are upgraded together with workspace object_store to 0.13.x.

Tests

• Unit Test — targeted cargo check (see below)
• Logic Test
• Benchmark Test
• No Test - Explain why

Validated with:

cargo check -p databend-common-storage
cargo check -p databend-common-storages-fuse
cargo check -p databend-common-sql
cargo check -p databend-common-storages-parquet

Lockfile: opendal 0.58.1 (direct), object_store_opendal 0.58.0, transitive opendal 0.55.0 still present via iceberg-rust.

Type of change

• Bug Fix (non-breaking change which fixes an issue)
• New Feature (non-breaking change which adds functionality)
• Breaking Change (fix or feature that could cause existing functionality not to work as expected)
• Documentation Update
• Refactoring
• Performance Improvement
• Other (please describe): dependency upgrade + API fallout fixes

AI assistance

• AI usage: Agent-assisted dependency bump, OpenDAL 0.54→0.58 API migration (custom layers, scheme/timestamp/http client rewrites), targeted compile validation, draft PR.
• Responsible human: @Xuanwo
• The responsible human has read every line of this diff and can explain each change

---

This change is 

[github-actions]

🤖 CI Job Analysis

Workflow: 30926357907

📊 Summary

• Total Jobs: 90
• Failed Jobs: 7
• Retryable: 0
• Code Issues: 7

❌ NO RETRY NEEDED

All failures appear to be code/test issues requiring manual fixes.

🔍 Job Details

• ❌ linux / test_compat_meta_query: Not retryable (Code/Test)
• ❌ linux / test_stateful_cluster: Not retryable (Code/Test)
• ❌ linux / test_stateful_standalone: Not retryable (Code/Test)
• ❌ linux / sqllogic / cluster (base, 2c, 2, http): Not retryable (Code/Test)
• ❌ linux / sqllogic / standalone (base, 2c, http): Not retryable (Code/Test)
• ❌ linux / sqllogic / cluster (base, 2c, 2, hybrid): Not retryable (Code/Test)
• ❌ linux / sqllogic / standalone (base, 2c, hybrid): Not retryable (Code/Test)

---

🤖 About

Automated analysis using job annotations to distinguish infrastructure issues (auto-retried) from code/test issues (manual fixes needed).

[Xuanwo]

object_store_opendal 0.58 requires object_store 0.13, while current deltalake (git rev) and datafuse-extras/lance still pin object_store 0.12. That leaves:

Any ideas? I'm not sure which version to use in this case.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 8d267e50a2

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".

[chatgpt-codex-connector]

Release permits after each raw I/O call

When max_concurrent_io_requests is enabled, acquire() stores the permit in self.permit but never clears it, so after a writer, lister, deleter, copier, or read stream performs its first operation, it occupies a slot until the entire handle is dropped. With a limit of 1, for example, an idle first writer blocks a second writer indefinitely rather than releasing capacity after write() completes; the assertion added at line 421 that the permit count returns to 1 after writer.write() will itself fail. Preserve the previous per-operation behavior by dropping/resetting the permit after each awaited raw I/O call.

AGENTS.md reference: AGENTS.md:L10-L11

Useful? React with 👍 / 👎.