[Feature]: Strip presentation styling from model-facing Mermaid diagrams

### Description

Two rule files spend tokens on Mermaid presentation styling that carries no information for the model:

1. common/process-overview.md (stated purpose: "AI model context loading") contains 20 style/linkStyle lines. The ALWAYS/CONDITIONAL status is already written in the node labels, so the colors are redundant for the model.
2. inception/workflow-planning.md Step 6 instructs the model to generate styled diagrams into execution-plan.md, including a styling rules block and color guidelines. Style syntax is a common source of the Mermaid parsing errors that common/content-validation.md exists to guard against, and the generated style block is reloaded as prior context by every later stage that loads the execution plan.

Proposal: remove the style/linkStyle lines from process-overview.md, and change workflow-planning.md Step 6 to request a plain flowchart with EXECUTE/SKIP/COMPLETED status written into node labels, with an explicit instruction not to emit style, classDef, or linkStyle lines.

### Use Case

Fewer tokens at rule load and at artifact generation, fewer Mermaid validation failures, and execution-plan.md stays readable in any renderer.

### Version

v1 (current)

### Area

Inception phase rules

### Additional Context

Reference implementation on my fork: https://github.com/Emillock/aidlc-workflows/tree/perf/slim-model-context (commit 8422afd).

Start by reading common/process-overview.md and inception/workflow-planning.md Step 6, then review common/content-validation.md for the Mermaid constraints mentioned in the issue. The work is complete when the process overview has no style/linkStyle lines and Step 6 requests plain flowcharts with status labels and no style, classDef, or linkStyle lines.

Written by the indexing model from the issue text.