Task: Compact leadership dashboard layout

Identity

Remove the entire “平台运行全景” top banner from /operations-dashboard so the leadership view starts with the indicator row.
Place compact, fixed-layout task-type and employee-operation rankings directly below the indicators.
Rank both dimensions by operation count and show operation count plus success count only; remove completion-rate presentation.
Place filter parameters and the task detail table below the rankings, including an explicit employee-name filter.
Preserve the existing leadership authorization boundary, business-safe task projection, and read-only drill-through behavior.

The page is a leadership operations view, not a technical audit or engineering diagnostics screen.
Visible ranking metrics must be counts, never completion percentages; the two ranking panels should remain visually stable even when their row counts differ.
A leader must be able to move from a KPI or ranking into the same filtered task list, then open one row to see the employee, original input, and final output.
Keep the current vanilla HTML/CSS/JavaScript stack and the current operations-dashboard API; do not redesign account authorization, modify task data, access ERP, or add an organization concept.
Do not restart or deploy the service without a separate explicit authorization.

Deleted the full top hero/banner and its live CSS, including the “平台运行全景” heading and explanatory copy. The first dashboard content is now the six-card indicator row.
Removed the prior time-trend block and rebuilt the next row as two equal, compact ranking panels: “任务类型排名” and “员工操作任务排名”. Both use a fixed 304px panel height with internal scrolling when needed.
Simplified every ranking row to ordinal, identity/type, operation count, and success count. No completion rate or percentage is calculated or rendered.
Moved the date presets, date inputs, employee name, task type, task result, and keyword query controls into the section immediately above the task details.
Replaced the prior task-flow cards with a semantic detail table containing operation time, employee name, task type, original input, final output, and task result. Rows remain keyboard- and pointer-openable for the complete read-only detail drawer.
Kept status KPI drill-through, task-type drill-through, employee drill-through, pagination, refresh, date ranges, business-safe input/output text, and the existing admin/team_lead visibility gate.
Updated cache-bust references and regression assertions for the compact dashboard presentation.
The already-running 8786 service serves these static page changes immediately. No service restart, deployment, database mutation, task mutation, ERP access, or external send was performed.

node --check LianSyn-platform/app.js: passed with the bundled Node runtime.
Focused account/dashboard regression: 8/8 passed.
node --run check:repo: 10/10 passed.
node --run check: passed.
node --run test:control-plane: 153/153 passed.
node --run test:legacy: 256/256 passed.
node --run build: passed.
git diff --check: passed.
Authenticated browser verification on http://127.0.0.1:8786/operations-dashboard?layout=compact-1 confirmed that the KPI row is first, the hero is absent, the two rankings are equal-height and compact, and only operation/success counts appear in ranking rows.
Browser data verification loaded 359 task operations, 251 successes, 23 task types, and 1 participating employee in the current 30-day range.
Browser interaction verification selected employee lwltadmin, kept the expected 359 matching operations, drilled the “删除订单” ranking into 43 matching operations, and opened a table row whose drawer displayed employee, original input, and final output.
Static presentation scan found no 平台运行全景, 完成率, completion-rate helper, or removed day-chart renderer in the live dashboard HTML/JavaScript.

The current data contains one employee, so the fixed employee ranking intentionally shows one row and unused panel space. It will populate without layout changes as additional ordinary users and team leads operate tasks.

Target: AUTH-001 and .project-docs/40-domain/business-rules.md during a later Integration Gate.
Proposal: record the accepted leadership-dashboard visual order as KPI indicators → compact task-type/employee rankings → filters and task detail table.
Proposal: task-type and employee rankings are count-based only and expose operation count plus success count; completion percentages are not part of leadership ranking semantics.
Evidence: LianSyn-platform/index.html, LianSyn-platform/app.js, LianSyn-platform/styles.css, control-plane/test/account-authorization.test.ts, control-plane/test/control-plane.test.ts, full regression checks, and authenticated browser verification summarized above.
Future impact: dashboard redesigns and regression tests should preserve the accepted information hierarchy, fixed ranking footprint, employee-name query, and business-language drill-through.
Semantic conflict: this refines the previous aggregate-first dashboard presentation by removing its hero and time trend; it does not change leadership access, task ownership, or API data semantics.
Human confirmation required: received in this task through the explicit requested layout and metric correction.