Support aggregate functions in Eval expressions

[LantaoJin]

Description

Aggregate functions could work in select clause even there is no group by as long as all items in select are aggregate functions.
Here are some examples:

select max(a) from t
select max(a), min(a), count(a) from t

But the following queries should throw exceptions

select a, max(a) from t
select a, max(a), min(a), count(a) from t
select *, max(a), min(a), count(a) from t

They could work with a group by

select a, max(a) from t group by b
select a, max(a), min(a), count(a) from t group by b
select *, max(a), min(a), count(a) from t group by b

Similar, aggregate functions in PPL could work in eval command, because an eval expression equals to add a projection to existing project list.
Here are some examples:

source=t | eval m = max(a) | fields m
source=t | eval m = max(a), n = count(a) | fields m, n
source=t | eval m = max(a) | n = count(a) | fields m, n

But the following PPL queries should throw exceptions

source=t | eval m = max(a) -- equals to all fields plus m
source=t | eval m = max(a), n = count(a) | fields a, m, n
source=t | eval m = max(a) | n = count(a) | fields a, m, n

Issues Resolved

#755

Check List

- [ ] Updated documentation (ppl-spark-integration/README.md)

• Implemented unit tests
• Implemented tests for combination with other commands
• Commits are signed per the DCO using --signoff

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[LantaoJin]

This change is not related. Just enable the test since fields exclude list has been supported.

[LantaoJin]

ditto