refactor(gui): drop Review page + normalization gate

Home is now the only entry point: the "Run analysis" button on the
upload section IS the review step (findings render inline via
render_findings_panel). Tool pages no longer gate on a passed
normalization — running the analyzer is sufficient context.

Removed:
- src/gui/pages/0_Review.py
- src/gui/components/gate.py (re-export seam)
- require_normalization_gate() in src/gui/components/_legacy.py
- "review" section enum in tools_registry.py
- Data Review entry in app.py navigation
- require_normalization_gate() calls + imports in all nine tool pages
- tests/gui/test_gate.py (whole file)
- TestReviewWorkflow in tests/gui/test_workflows.py
- 0_Review entry in tests/gui/test_smoke.py PAGE_SLUGS
- stash_upload's normalization_result+normalization_for stashing
- stash_upload_without_gate (was the gate's negative-path helper)

2017 tests pass (16 retired with the gate flow).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/gui/tools_registry.py

@@ -22,10 +22,9 @@ from typing import Literal
 
 Tier = Literal["core", "pro", "enterprise"]
 Status = Literal["Ready", "Coming Soon"]
-# Sidebar grouping. The Review gate is its own section; cleaners,
-# transformations, and automations group the tools by what the user is
-# trying to accomplish rather than by implementation detail.
-Section = Literal["review", "cleaners", "transformations", "automations"]
+# Sidebar grouping. Tools are bucketed by what the user is trying to
+# accomplish rather than by implementation detail.
+Section = Literal["cleaners", "transformations", "automations"]
 
 
 @dataclass(frozen=True)
@@ -152,7 +151,6 @@ TOOLS: list[Tool] = [
 # Display labels for each sidebar section. Kept here so i18n falls back
 # to a sensible English string if a translation pack is missing the key.
 SECTION_LABELS: dict[Section, str] = {
-    "review": "Data Review",
     "cleaners": "Data Cleaners",
     "transformations": "Transformations",
     "automations": "Automations",