Optimize CLI output defaults for agent efficiency (#94)

[Dimwiddle]

Description

Implements issue #94 end-to-end by centralizing CLI output policy and applying compact/TTY-aware defaults across command surfaces.

Key outcomes:

• Added shared output policy helpers for TTY-aware format defaults, compact JSON serialization, and --pretty support.
• Switched command --format defaults to auto-detect (table for TTY, json for non-TTY) while preserving explicit override behavior.
• Added status --verbose; default JSON status now returns summary-first payloads.
• Added compact-mode behavior (SPECLEFT_COMPACT=1) with next --limit 1 default.
• Reduced success payload verbosity for key write/validation flows (features validate, features add, features add-scenario, test skeleton, test stub, init, threshold-based coverage).
• Added deterministic fix_command fields for applicable failure outputs in doctor, contract test, and enforce JSON payloads.
• Updated skill docs/template to use SPECLEFT_COMPACT=1 and exit-code-first workflow guidance.
• Updated command and acceptance tests for new output defaults and payload contracts.

Type of Change

• Refactor
• Feature
• Documentation update
• Test updates

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my code
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

Testing

• make lint -- passing
• pytest -- 448 passed

Related Issues

Closes #94