docs: add "Bring your data" step to homepage journey

[saadqbal]

Summary

• Adds an explicit "Bring your data" step to the homepage's "How It Works" between workspace creation and use-case definition.
• Rewords Step 2 from "Select datasets, set evaluation metrics" to "Pick from your prepared datasets, set evaluation metrics" so it doesn't read as a UI action performed during use-case creation.
• Same rewording on the matching "Next Steps" card.

The homepage previously inverted the actual journey: it jumped from "Create your AI workspace" straight to "Define a use case", with datasets selected as a sub-action of that step. That contradicted /create-use-case (which correctly lists ingestion as the first workflow step) and sent newcomers into the platform looking for a dataset selector with no datasets to select.

Cross-references the same bug class in tracebloc/data-ingestors#66 (Readme prereq listed "active use case" as required before ingest).

Closes #28

Diff

   <Step title="Create your AI workspace">…</Step>
+  <Step title="Bring your data">
+    Validate and stage your datasets on your cluster. Metadata syncs to the web app — raw data stays put.
+  </Step>
   <Step title="Define a use case">
-    Select datasets, set evaluation metrics. Use a template or build your own.
+    Pick from your prepared datasets, set evaluation metrics. Use a template or build your own.
   </Step>

Test plan

• Mintlify preview renders the 6-step <Steps> block correctly (was 5).
• "Next Steps" card wording matches the rest of the homepage tone.
• Homepage journey now matches /create-use-case ordering.

🤖 Generated with Claude Code

---

Note

Low Risk
Low risk documentation-only update that reorders and rewords homepage guidance; the main risk is minor messaging/UX confusion if the new step copy is inaccurate.

Overview
Updates the overview/tracebloc.mdx homepage journey to explicitly include a “Bring your data” step between workspace setup and use-case definition, clarifying that datasets are validated/staged on the user’s cluster and only metadata syncs.

Rewords related copy in the hero and “Next Steps” cards and the “Define a use case” step to emphasize picking from prepared datasets rather than selecting datasets during use-case creation.

^{Reviewed by Cursor Bugbot for commit a5bd224. Bugbot is set up for automated code reviews on this repo. Configure here.}

[LukasWodka]

👋 Heads-up — Code review queue is at 14 / 8

Above the WIP limit. The team convention is to review existing PRs before opening new work.

Open PRs currently in Code review (oldest first):

• backend#640 — security: harden tracebloc_model_validation Bandit plugin (TBT001) · author: @aptracebloc · reviewer: @saqlainsyed007
• docs#14 — docs: migrate join-use-case API examples to snake_case (0.7.0) · author: @divyasinghds · no reviewer assigned
• docs#26 — docs: add "How training works" page · author: @divyasinghds · no reviewer assigned
• docs#27 — docs: add automated upstream sync workflow · author: @divyasinghds · no reviewer assigned
• docs#4 — docs: switch install to extras syntax for tracebloc_package 0.6.33 · author: @LukasWodka · reviewer: @saadqbal
• tracebloc-client#244 — feat(object_detection): default color augmentation from augmentation_config · author: @divyasinghds · no reviewer assigned
• tracebloc-client#245 — feat(semantic_segmentation): default mask-aware augmentation from augmentation_config · author: @divyasinghds · no reviewer assigned
• tracebloc-client#246 — feat(keypoint_detection): default keypoint-aware augmentation from augmentation_config · author: @divyasinghds · no reviewer assigned
• tracebloc-client#247 — fix(augmentation): map flat SDK keys into augmentation_config · author: @divyasinghds · no reviewer assigned
• tracebloc-client#248 — fix(keypoint_detection): repair cycle PCK threshold and R-CNN cycle metrics · author: @divyasinghds · no reviewer assigned

Pull from review before opening new work. (This is a nudge from the kanban WIP check, not a block.)