only run regex replaces on DBs with UTF8 encoding #8278

[pdurbin]

What this PR does / why we need it:

Deployment fails if the encoding is SQL_ASCII. Here we only run the regex replaces if the database is UTF8.

Which issue(s) this PR closes:

Closes #8278

Special notes for your reviewer:

Devs will need to delete the old "V5.8.0.2__8018-invalid-characters.sql" SQL migration from their flyway table because the checksum is changing.

Suggestions on how to test this:

Make sure the replaces still happen

Does this PR introduce a user interface change? If mockups are available, please link/include them here:

No.

Is there a release notes update needed for this change?:

No.

Additional documentation:

No

[scolapasta]

@pdurbin For the PR, you should be able to just use "replace" function instead of reg_exp replace and not need all the extra logic checking encoding.

[pdurbin]

@scolapasta I tried just replace (please see below) and I got the same error:

dvndb=# UPDATE datasetfieldvalue SET value = replace(value, U&'\FFFE', '', 'g');
ERROR:  conversion between UTF8 and SQL_ASCII is not supported
LINE 1: ...E datasetfieldvalue SET value = replace(value, U&'\FFFE', ''...
                                                             ^
dvndb=# \l
                             List of databases
   Name    |  Owner   | Encoding  | Collate | Ctype |   Access privileges   
-----------+----------+-----------+---------+-------+-----------------------
 dvndb     | dvnapp   | SQL_ASCII | C       | C     |