[wb_acq_core] On trigger acquisitions we can have a sample not belonging to the correct transaction

[lerwys]

Problem: On trigger acquisitions, for some unknown reason, we could have the signal "fifo_fc_mux_cnt" on acq_fc_fifo module ending in something different than 0 at the end of a transaction.

This causes the the first sample of the next transaction to have atoms belonging to a previous transaction.

This could be detected in simulation, but our testbench was incomplete and not checking this issue properly. The testbench is comparing the output of acq_fc_fifo with the data read from DDR. So, problems arising before the acq_fc_fifo or in itself would not be detected!

Roadmap:

1. Fix testbench to correctly validates our module (comparing input data with the data read from DDR)
2. Investigate issue to why fifo_fc_mux_cnt signal ends in something different than 0 at the end of some transactions.

[lerwys]

Commit 8ffd615 (fixing trigger alignement) did not solve the error completely, although it improved the testbench completion