fix(seq): handle 0e... scientific notation without padding

[karanabe]

Fix: #6926

This pull request fixes the issue where seq mishandles scientific notation in the form of 0e... by treating it as zero.

Note;
While testing with bash util/run-gnu-test.sh, I discovered another bug unrelated to this fix. I will file a new issue.

cargo run -q seq 0x1p-1 2
seq: invalid hexadecimal argument: '0x1p-1'

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[cakebaker]

I don't know if you have seen it: the changes break numberparse::tests::test_parse_big_int:

---- numberparse::tests::test_parse_big_int stdout ----
thread 'numberparse::tests::test_parse_big_int' panicked at src/uu/seq/src/numberparse.rs:402:9:
assertion `left == right` failed
  left: BigDecimal(BigDecimal(sign=NoSign, scale=0, digits=[]))
 right: BigDecimal(BigDecimal(sign=Plus, scale=0, digits=[1]))

[karanabe]

Sorry. I was looking at the integration test, but I must have missed the unit test. I fixed it.

[cakebaker]

Sorry. I was looking at the integration test, but I must have missed the unit test. I fixed it.

No problem :) Can you please run cargo fmt?

[karanabe]

Thank you for the attention. I seem to have gotten impatient, I didn't follow the DEVELOPMENT.md procedure.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[cakebaker]

Thanks for your PR!