TPC-H Query 1 has regressed

[avantgardnerio]

Describe the bug

When running TPC-H query 1, this is the result:

2022-10-05 14:42:16,272 [2560011]   WARN - #c.i.d.d.BaseDatabaseErrorHandler$UnknownErrorInfo - Error while executing SQL "select l_returnflag, l_linestatus, sum(l_quantity) as sum_qty, sum(l_extendedprice) as sum_base_price,
    sum(l_extendedprice * (1 - l_discount)) as sum_disc_price,
    sum(l_extendedprice * (1 - l_discount) * (1 + l_tax)) as sum_charge, avg(l_quantity) as avg_qty,
    avg(l_extendedprice) as avg_price, avg(l_discount) as avg_disc, count(*) as count_order
from lineitem
where l_shipdate <= date '1998-09-02'
group by l_returnflag, l_linestatus
order by l_returnflag, l_linestatus": Error executing plan: Error planning job l6OLwA0: DataFusionError(Internal("The type of Timestamp(Nanosecond, None) <= Date32 of binary physical should be same"))
java.sql.SQLException: Error while executing SQL "select l_returnflag, l_linestatus, sum(l_quantity) as sum_qty, sum(l_extendedprice) as sum_base_price,
    sum(l_extendedprice * (1 - l_discount)) as sum_disc_price,
    sum(l_extendedprice * (1 - l_discount) * (1 + l_tax)) as sum_charge, avg(l_quantity) as avg_qty,
    avg(l_extendedprice) as avg_price, avg(l_discount) as avg_disc, count(*) as count_order
from lineitem
where l_shipdate <= date '1998-09-02'
group by l_returnflag, l_linestatus
order by l_returnflag, l_linestatus": Error executing plan: Error planning job l6OLwA0: DataFusionError(Internal("The type of Timestamp(Nanosecond, None) <= Date32 of binary physical should be same"))

To Reproduce

Run TPC-H query 1.

Expected behavior

A result is returned.

[alamb]

We have fixed a few bugs in this coercion area recently -- it "might" work if you try with latest master.

Now that we have the optimizer integration test, it would be great to at least get tests for all tpch querries into that

[avantgardnerio]

I need to look into this again - @andygrove said he ran it and it worked. I think my schema inference may have picked the wrong data type for this timestamp and gotten different results.

[andygrove]

Could be related to #3764

[alamb]

@andygrove has proposed an improvement to testing #3790

[Dandandan]

tests::run_q1 is working on master (just confirmed). Hopefully when the TPC-H regression tests are there we can avoid errors creeping in.