Add incident reports for guard-triggered traces

[bmdhodl]

Summary

• add a dependency-free incident reporting module for guard-triggered traces
• extend the CLI with agentguard incident and agentguard report --format markdown|html|json
• document the incident-report flow in the README, getting-started guide, and cost guardrail example

Why

This turns guard trips into a product-led proof-of-value moment: local traces can now become shareable incident summaries with conservative savings estimates and a clear upgrade path to the hosted dashboard.

Testing

• python -m ruff check sdk/agentguard/
• python -m pytest sdk/tests/test_architecture.py -v
• python -m bandit -r sdk/agentguard/ -s B101,B110,B112,B311 -q
• python -m pytest tests/test_reporting.py tests/test_cli_report.py -v (from sdk/)

Full suite note

python -m pytest sdk/tests/ -v --cov=agentguard --cov-report=term-missing --cov-fail-under=80 reached 508 passes with 2 unrelated failures on this Windows environment:

• sdk/tests/test_init.py::TestInit::test_init_custom_trace_file
• sdk/tests/test_tracing.py::test_trace_emits_events

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6e5789bf45

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".