[fix](meger-on-write) fix query result wrong when schema change

[liaoxin01]

Proposed changes

Issue Number: close #xxx

Problem：

1. failed to initialize storage reader. tablet=40321.1191259674.1f4189c7184f58ad-15f3320801a27cb0, res=[INTERNAL_ERROR]fail to find path in version_graph. spec_version: 0-45
2. duplicate key in result set

Why：

1. when tablet in converting，we don't commit the rowset just return it. So the rowset cann't be publushed, we can't find it in the version graph. this was introduced by: [improvement](merge-on-write) move segment check to delta writer #18643

        if (_tablet->tablet_state() == TABLET_NOTREADY &&
            SchemaChangeHandler::tablet_in_converting(_tablet->tablet_id())) {
            return Status::OK();
        }

2. when tablet in converting, we should clear _rowset_ids. Otherwise, we will calculate the delete bitmap incorrectly when publish.

Further comments

If this is a relatively large or complex change, kick off the discussion at dev@doris.apache.org by explaining why you chose the solution you did and what alternatives you considered, etc...

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[dataroaring]

LGTM

[dataroaring]

run buildall

[github-actions]

PR approved by at least one committer and no changes requested.

[github-actions]

PR approved by anyone and no changes requested.

[hello-stephen]

(From new machine)TeamCity pipeline, clickbench performance test result:
the sum of best hot time: 45 seconds
stream load tsv: 510 seconds loaded 74807831229 Bytes, about 139 MB/s
stream load json: 20 seconds loaded 2358488459 Bytes, about 112 MB/s
stream load orc: 65 seconds loaded 1101869774 Bytes, about 16 MB/s
stream load parquet: 31 seconds loaded 861443392 Bytes, about 26 MB/s
insert into select: 29.5 seconds inserted 10000000 Rows, about 338K ops/s
storage size: 17162507721 Bytes

[Yukang-Lian]

LGTM

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[liaoxin01]

run buildall

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[hello-stephen]

(From new machine)TeamCity pipeline, clickbench performance test result:
the sum of best hot time: 45.19 seconds
stream load tsv: 506 seconds loaded 74807831229 Bytes, about 140 MB/s
stream load json: 20 seconds loaded 2358488459 Bytes, about 112 MB/s
stream load orc: 65 seconds loaded 1101869774 Bytes, about 16 MB/s
stream load parquet: 31 seconds loaded 861443392 Bytes, about 26 MB/s
insert into select: 29.1 seconds inserted 10000000 Rows, about 343K ops/s
storage size: 17169253700 Bytes

[zhannngchen]

LGTM

[github-actions]

PR approved by at least one committer and no changes requested.