BUG: handle overflow during exponent parsing in read_csv

[Alvaro-Kothe]

• closes BUG: incorrect string parsing leads to segfault #62617
• closes BUG: read_csv (c engine, round_trip precision) platform inconsistency with large exponents #38794
• closes BUG: read_csv is inconsistent with large exponent #62740
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

---

This patch uses strtol to compute the exponent and checks overflow by the errno assigned on strtol and with checked_add.

I deleted the previous tests regarding large exponents because they were validating the undefined behavior caused by the overflow, where in one test asserts that 10E-99999999999999999 is parsed as 0.0, while on the other 10E-999999999999999999 (contains 1 digit more) is parsed as a string.

Now, the float parsing on overflow follows the results from pyarrow.

[WillAyd]

Great thanks @Alvaro-Kothe