Avoid uint64_t overflow in Decimal::operator/() and fix Decimal::ceil()

Avoid uint64_t overflow in Decimal::operator/()

This patch changes Decimal::operator/() not to cause uint64_t overflow in
multiplication and addition during calculating quotient. Original code
wrongly assumed each division loop generate at most two digits, by
|MaxCoefficient < 100|, however this assumption is wrong such as
50,000 / 99,9999.

This patch also fixes assertion failure |n> Precision|, where |Precision| == 18, in |scaleUp(x, n)| via |Decimal::ceil()|. Before this patch, we don't have 18 digits quotient with 10^-18 exponent from result of division operator.

BUG=463510
TEST=blink_platform_unittests --gtest_filter=DecimalTest.Division

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=191294

[yosin_UTC9, 2015-03-04 04:14:38]

yosin@chromium.org changed reviewers:
+ tkent@chromium.org

[tkent, 2015-03-04 04:40:03]

lgtm

[tkent, 2015-03-04 16:20:09]

The CQ bit was checked by tkent@chromium.org

[tkent, 2015-03-04 16:20:10]

The patchset sent to the CQ was uploaded after l-g-t-m from tkent@chromium.org
Link to the patchset: https://codereview.chromium.org/977593004/#ps20001 (title:
"2015-03-04T15:05:41")

[commit-bot: I haz the power, 2015-03-04 16:20:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/977593004/20001

[commit-bot: I haz the power, 2015-03-04 17:20:31]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=191294