Redesign the interactive review and fix AI explanations

[VIVAAN-DHAWAN]

Summary

• Centres the loading transition, section picker, and focused review output in the terminal.
• Rewrites the guided prompts and error messages with clearer next steps.
• Keeps generated AI explanations in the interactive menu; the previous duplicate menu setup silently discarded that section.
• Does not ask to enable AI explanations when no API key is configured, and explains how to configure one.
• Replaces opaque GitHub 404 output with an actionable pull-request-not-found message.

Behaviour preserved

Commands, deterministic verdicts, keyboard bindings, report data, and the existing logo asset remain unchanged.

Verification

• uv run --extra dev pytest -q (49 passed)
• uv run python -m compileall -q src
• git diff --check

[github-actions]

Diffly verdict: QUARANTINE

#11 · 2 files · 232 lines changed · checks: PENDING

Risk flags: NO_TEST_COVERAGE (medium), CHECKS_PENDING (low)

Risk flags and reasoning

• NO_TEST_COVERAGE (medium): Changed production files have no obvious neighboring or repository test coverage. — src/diffly_cli/cli.py
• CHECKS_PENDING (low): Required status checks are still pending. — combined commit status; diffly; test (3.10); test (3.11); test (3.12); test (3.13)
• QUARANTINE because at least one changed production file lacks obvious test coverage.
• QUARANTINE because required checks are still running.

Blast-radius summary

File	Change	Symbols	Direct callers	Tests
src/diffly_cli/cli.py	+137/-47	_loading_mark, interactive_sections, interactive_view, show_loading_screen	_loading_mark, interactive_sections, show_loading_screen	—
tests/test_regressions.py	+48/-0	<top-level changes>	—	—

Deterministic triage is authoritative; any optional LLM explanation cannot change the verdict.