Fix SweepSummary status inference never running due to "good" default

[Copilot]

SweepSummary.__post_init__() was intended to infer scaling_status/latency_status from verdict text when not explicitly provided. However, both fields defaulted to "good", which is itself a valid status — so the not in valid_statuses guard was always False and inference never ran. get_overall_status() consequently always returned "good".

Changes

• SweepSummary: Change scaling_status and latency_status defaults from "good" to None (Optional[str]), so __post_init__ inference always runs unless the caller explicitly passes a valid status string.
• __post_init__ docstring: Updated to reflect the corrected semantics — infer unless overridden.

# Before: "good" is in valid_statuses → inference branches never execute
scaling_status: str = "good"
latency_status: str = "good"

# After: None is not in valid_statuses → inference always runs
scaling_status: Optional[str] = None
latency_status: Optional[str] = None

Explicit caller overrides (scaling_status="warning") are still respected since "warning" passes the not in valid_statuses check.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.