cxgbe(4): Fixes to tx coalescing.

• The behavior implemented in r362905 resulted in delayed transmission of packets in some cases, causing performance issues. Use a different heuristic to predict tx requests.

• Add a tunable/sysctl (hw.cxgbe.tx_coalesce) to disable tx coalescing entirely. It can be changed at any time. There is no change in default behavior.

(cherry picked from commit 3447df8bc5b342bd88d565641435284ff620ee2a)