fix(scripts): stop treating command failure as a pattern match

[mkoura]

Three related ways a shell test reported the wrong answer, all by letting a producer's exit status stand in for -- or be mistaken for -- the result of the match.

grep -q exits on first match and closes the pipe while the producer may still have a pending write. The producer dies on SIGPIPE with 141, and pipefail promotes that to the pipeline's exit status, so the test reads as "no match" even when the pattern is present. It is a race on syscall interleaving, not a size threshold, so it fires intermittently and the odds worsen as the scanned output grows. The three _wait_for_tx_*_tx helpers hung a full 7200s timeout each, then failed cluster start, because a healthy tx tool fills the last 100 log lines with the very pattern being matched -- so the match lands on line 1, leaving tail with the most left to write. The healthier the tool, the likelier the false negative. Keep the producers out of the pipelines.

check_spend_success folded the query and the match into one condition, so a failing query utxo produced no output, no lovelace match, and the negation returned 0 -- inputs reported as spent, retry loop skipped. Reachable both through the SIGPIPE above and through any ordinary query error. Capture the query and test its status separately.

grep -c exits 1 when it counts zero, so the CC registration check aborted at the ERR trap instead of reaching its own error message.