would it be reasonable to submit a RFE to check for type mis-matches for SQL precompiler?

Applying the DB2i group ptf 13, which included PTF SI76030 (7.3ptf
SI76035), will cause SQLRPGLE programs to start issuing SQLSTATE:
'22023'/-302 or '53045'/-678 when you have an invalid comparison between a
numeric value and blank(s) or a numeric value and an empty string.

We did indeed have a few programs (so far we have caught 3 programs) that
had invalid comparisons, and we can't find an easy way to search for
existing cases of these invalid conditions.

or cases included (field names changed for example sake):
1) COALESCE(numericSEQ#,' ' ) // coalesce on field from left join

2) CASE When numericdate = 20391231 then int(current date + 7
days)
Else ' ' END as numericdate2

Would it be a reasonable request to have the compiler catch these mistakes?