[FLINK-39566][runtime-web] Add checkpoint duration Gantt view

[spuru9]

What is the purpose of the change

Adds a Gantt chart view to the job's Checkpoints tab so operators can see at a glance why a checkpoint was slow. The same information already exists in the per-subtask table, but it's spread across many rows of numbers; this view makes stragglers and stuck phases visually obvious instead of requiring a manual scan.

Brief change log

• New Gantt tab on the job Checkpoints page (frontend only, no backend changes)
• Recent Checkpoints strip: last 60 checkpoints as colored bars (completed / savepoint / in-progress / failed); click a bar to drill in
• Per-checkpoint Gantt: one row per subtask with stacked bars for the four checkpoint phases, sorted by duration so the slowest subtasks rise to the top
• Pin / Follow newest toggle so a chosen checkpoint stays put for analysis, and PNG export for incident reports
• Auto-refresh aligned to checkpoint cadence
• Suppresses the global error toast for the per-checkpoint details endpoint while a checkpoint is still in progress (it returns 404 until ack), so users don't see spurious notifications

Verifying this change

This change is frontend-only and is verified manually:

• Run a Flink cluster with a streaming job that has periodic checkpoints
• Open the job's Checkpoints tab and switch to the new Gantt view
• Confirm the recent-checkpoint strip populates and updates as new checkpoints complete
• Click a bar — the per-subtask Gantt renders for that checkpoint and the view stays pinned
• Click Follow newest — auto-tracking resumes
• Click Export PNG — image downloads
• Trigger a savepoint and a failed checkpoint and confirm they render with the right colors

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): no (uses @antv/g2, already a project dependency)
• The public API, i.e., is any changed class annotated with @Public(Evolving): no
• The serializers: no
• The runtime per-record code paths (performance sensitive): no
• Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn, ZooKeeper: no
• The S3 file system connector: no

Documentation

• Does this pull request introduce a new feature? yes
• If yes, how is the feature documented? not applicable (UI-only feature, discoverable from the Checkpoints tab)

---

Was generative AI tooling used to co-author this PR?

• Yes (Claude Code)

[flinkbot]

CI report:

• a66afce Azure: FAILURE

Bot commands

The @flinkbot bot supports the following commands:

• @flinkbot run azure re-run the last Azure build