The problem is that in case of incorrect shutdown, i.e.:

    2025.04.03 22:42:03.620982 [ 14841 ] {632881cd-4918-414c-a2ed-27bb6beee664} <Error> executeQuery: Code: 219. DB::Exception: New table appeared in database being dropped or detached. Try again. (DATABASE_NOT_EMPTY) (version 25.4.1.1) (from [::1]:37582) (comment: 03258_refreshable_mv_misc.sh) (query 1, line 2) (in query: drop database test_15_03258;), Stack trace (when copying this message, always include the lines below):

    0. ./contrib/llvm-project/libcxx/include/__exception/exception.h:113: Poco::Exception::Exception(String const&, int) @ 0x0000000020b619a0
    1. ./ci/tmp/build/./src/Common/Exception.cpp:108: DB::Exception::Exception(DB::Exception::MessageMasked&&, int, bool) @ 0x0000000010bd82d4
    2. DB::Exception::Exception(PreformattedMessage&&, int) @ 0x00000000086edcc0
    3. DB::Exception::Exception<>(int, FormatStringHelperImpl<>) @ 0x00000000086fc67a
    4. ./ci/tmp/build/./src/Interpreters/DatabaseCatalog.cpp:615: DB::DatabaseCatalog::detachDatabase(std::shared_ptr<DB::Context const>, String const&, bool, bool) @ 0x00000000180a72d7
    5. ./ci/tmp/build/./src/Interpreters/InterpreterDropQuery.cpp:535: DB::InterpreterDropQuery::executeToDatabaseImpl(DB::ASTDropQuery const&, std::shared_ptr<DB::IDatabase>&, std::vector<StrongTypedef<wide::integer<128ul, unsigned int>, DB::UUIDTag>, std::allocator<StrongTypedef<wide::integer<128ul, unsigned int>, DB::UUIDTag>>>&) @ 0x00000000186c1105
    6. ./ci/tmp/build/./src/Interpreters/InterpreterDropQuery.cpp:364: DB::InterpreterDropQuery::executeToDatabase(DB::ASTDropQuery const&) @ 0x00000000186bbe78
    7. ./ci/tmp/build/./src/Interpreters/InterpreterDropQuery.cpp:100: DB::InterpreterDropQuery::executeSingleDropQuery(std::shared_ptr<DB::IAST> const&) @ 0x00000000186bae97
    8. ./ci/tmp/build/./src/Interpreters/InterpreterDropQuery.cpp:73: DB::InterpreterDropQuery::execute() @ 0x00000000186baab0
    9. ./ci/tmp/build/./src/Interpreters/executeQuery.cpp:1458: DB::executeQueryImpl(char const*, char const*, std::shared_ptr<DB::Context>, DB::QueryFlags, DB::QueryProcessingStage::Enum, DB::ReadBuffer*, std::shared_ptr<DB::IAST>&) @ 0x0000000018be45da
    10. ./ci/tmp/build/./src/Interpreters/executeQuery.cpp:1625: DB::executeQuery(String const&, std::shared_ptr<DB::Context>, DB::QueryFlags, DB::QueryProcessingStage::Enum) @ 0x0000000018bdf1ef
    11. ./ci/tmp/build/./src/Server/TCPHandler.cpp:665: DB::TCPHandler::runImpl() @ 0x000000001b9cb665
    12. ./ci/tmp/build/./src/Server/TCPHandler.cpp:2630: DB::TCPHandler::run() @ 0x000000001b9f23c8
    13. ./ci/tmp/build/./base/poco/Net/src/TCPServerConnection.cpp:40: Poco::Net::TCPServerConnection::start() @ 0x0000000020c71da3
    14. ./ci/tmp/build/./base/poco/Net/src/TCPServerDispatcher.cpp:115: Poco::Net::TCPServerDispatcher::run() @ 0x0000000020c72612
    15. ./ci/tmp/build/./base/poco/Foundation/src/ThreadPool.cpp:205: Poco::PooledThread::run() @ 0x0000000020be7cc3
    16. ./ci/tmp/build/./base/poco/Foundation/src/Thread.cpp:45: Poco::(anonymous namespace)::RunnableHolder::run() @ 0x0000000020be6070
    17. ./base/poco/Foundation/src/Thread_POSIX.cpp:335: Poco::ThreadImpl::runnableEntry(void*) @ 0x0000000020be442a
    18. __tsan_thread_start_func @ 0x0000000008661428
    19. ? @ 0x00007f396f9fdac3
    20. ? @ 0x00007f396fa8f850

The table will still exist, but the view will be NULL due to shutdown()
had been called already.

And after this any ALTER of it will lead to crash.

So fix this by obtaining info for notifying under lock

P.S. I've looked through other places and it is either called under
refreshTask() (which should not be called after shutdown()) or already
has check for `view` under mutex.

Fixes: ClickHouse#78103
v2: do not call RefreshSet::notifyDependents() under lock (will lead to lock order inversion, found by @al13n321)