Update feedInTask tests to reflect portRounds-based progress tracking

[Copilot]

The feedInTask struct dropped the singular round field in favor of rounds (total) and portRounds (per-port slice) to enable opportunistic, barrier-free data feeding. Existing tests still referenced the old round field and didn't cover the new semantics.

Changes

• Removed stale round field from task initialization in "should do feed in" test; replaced with rounds and portRounds
• Added assertions to "should handle FeedIn/Collect API" tests verifying correct initialization of rounds and portRounds
• Added layout validation tests — validateTaskLayout panics on non-stride-multiple data lengths or stride < port count
• Added opportunistic feed-in test — port 2 blocked on tick 1, advances independently; verifies per-port portRounds state after each tick and final data ordering
• Added opportunistic collect test — mirrors feed-in: port 2 receives late, verifies per-port portRounds progression and final collected data

// Task now tracks progress per port, not globally
driver.feedInTasks[sideIndex] = []*feedInTask{{
    data:       []uint32{1, 2, 3, 4, 5, 6},
    rounds:     2,
    portRounds: []int{0, 0, 0}, // each port advances independently
    ...
}}
// After tick 1 with port 2 blocked:
Expect(task.portRounds).To(Equal([]int{1, 0, 1}))

---

💡 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.