Change the remaning columns that has text type in mysql,mariadb to longtext

[qnkhuat]

Fixes #26212 and is a follow up of #18749

Sanity check

SELECT table_schema, table_name, column_name, data_type
FROM INFORMATION_SCHEMA.COLUMNS
WHERE (data_type='text' OR data_type='blob') AND table_schema='metabase';

-- before
+--------------+--------------------------------------+-------------+-----------+
| TABLE_SCHEMA | TABLE_NAME                           | COLUMN_NAME | DATA_TYPE |
+--------------+--------------------------------------+-------------+-----------+
| metabase     | collection_permission_graph_revision | after       | text      |
| metabase     | collection_permission_graph_revision | before      | text      |
| metabase     | collection_permission_graph_revision | remark      | text      |
| metabase     | permissions_revision                 | after       | text      |
| metabase     | permissions_revision                 | before      | text      |
| metabase     | permissions_revision                 | remark      | text      |
| metabase     | QRTZ_BLOB_TRIGGERS                   | BLOB_DATA   | blob      |
| metabase     | QRTZ_CALENDARS                       | CALENDAR    | blob      |
| metabase     | QRTZ_JOB_DETAILS                     | JOB_DATA    | blob      |
| metabase     | QRTZ_TRIGGERS                        | JOB_DATA    | blob      |
| metabase     | secret                               | value       | blob      |
+--------------+--------------------------------------+-------------+-----------+

-- after
+--------------+--------------------+-------------+-----------+
| TABLE_SCHEMA | TABLE_NAME         | COLUMN_NAME | DATA_TYPE |
+--------------+--------------------+-------------+-----------+
| metabase     | QRTZ_BLOB_TRIGGERS | BLOB_DATA   | blob      |
| metabase     | QRTZ_CALENDARS     | CALENDAR    | blob      |
| metabase     | QRTZ_JOB_DETAILS   | JOB_DATA    | blob      |
| metabase     | QRTZ_TRIGGERS      | JOB_DATA    | blob      |
+--------------+--------------------+-------------+-----------+

[codecov]

Codecov Report

Base: 64.28% // Head: 64.28% // Increases project coverage by +0.00% 🎉

Coverage data is based on head (def7695) compared to base (cc589cf).
Patch has no changes to coverable lines.

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #26223   +/-   ##
=======================================
  Coverage   64.28%   64.28%           
=======================================
  Files        3148     3148           
  Lines       92202    92202           
  Branches    11699    11699           
=======================================
+ Hits        59269    59270    +1     
+ Misses      28243    28242    -1     
  Partials     4690     4690           

Flag	Coverage Δ
back-end	85.56% <ø> (+<0.01%)	⬆️
front-end	45.13% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...etabase_enterprise/serialization/v2/utils/yaml.clj	100.00% <0.00%> (+2.12%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[calherries]

Do we really need this test? It is relatively costly to execute, and we don't run tests like this for all migrations. I feel the same way about convert-text-to-longtext-migration-test actually.

I can see the need to test more complicated migrations that involve SQL like v45.00-049 but in this case I think your sense check in the PR description is sufficient.

[qnkhuat]

Fair point, I'll remove this test.

[calherries]

LGTM, approving even if I question whether the test can be deleted.