Add fault correlation/muting capability to FaultManager to handle "expected cascades" of faults.

[mfaferek93]

Summary

Add fault correlation/muting capability to FaultManager to handle "expected cascades" of faults.

Use cases (from this discussion: https://discordapp.com/channels/1451323858547118216/1459576596733104158):

• e-Stop cascades: when e-stop triggers, downstream motor/comm faults are expected noise
• OTA/restart transitions: temporary bursts of timeouts and comm errors during updates
• degradation modes: intentionally disabled subsystems shouldn't keep surfacing known faults

Currently all faults are treated equally, making it hard to identify the actual root cause when multiple related faults fire simultaneously.

---

Proposed solution (optional)

Root Cause and Symptoms mapping

Instead of simple muting, define relationships between faults:

# 1. Define fault patterns (regex matching on diagnostic name/message)
fault_patterns:
  motor_low_power:
    name: "Motor.*"
    message: "Low Voltage|Power Loss"

  motor_comm_timeout:
    name: "Motor.*"
    message: "Timeout|No Response"

# 2. Define root causes with expected symptoms
root_causes:
  estop_pressed:
    name: "E-Stop Pressed"
    symptoms:
      - motor_low_power
      - motor_comm_timeout

Benefits over simple muting:

• Faults aren't hidden, they're contextualized
• Unexpected faults (not matching any symptom) stand out as potentially real issues

Implementation location: FaultManager (centralized, full system context)

---

Additional context (optional)

[nnarain]

lgtm! thanks for looping me into the discussion. I'll look into experimenting with this project in our systems at some point.