feat: add QUERY_TECHNICAL_MARK to all UI queries

[artemmufazalov]

Closes #1490

Why to mark queries?

1. UI request to system tables won't be displayed in queries tables
2. It will be easier to distinguish UI queries from users' queries in case of incidents

Also I added database to all queries, so requests will use absolute paths and work in nested databases. Renamed path to database in some places for clarity

CI Results

Test Status: ✅ PASSED

📊 Full Report

Total	Passed	Failed	Flaky	Skipped
264	263	0	0	1

Test Changes Summary ⏭️1

⏭️ Skipped Tests (1)

1. Streaming query shows some results and banner when stop button is clicked (tenant/queryEditor/queryEditor.test.ts)

Bundle Size: ✅

Current: 80.81 MB | Main: 80.81 MB
Diff: 0.01 KB (-0.00%)

✅ Bundle size unchanged.

ℹ️ CI Information

• Test recordings for failed tests are available in the full report.
• Bundle size is measured for the entire 'dist' directory.
• 📊 indicates links to detailed reports.
• 🔺 indicates increase, 🔽 decrease, and ✅ no change in bundle size.

[astandrik]

If database is empty (as far as I can see - it can be '' because we are checking its value in ternary above) - will /.sys/... be okay as select source?

[artemmufazalov]

.sys will work, /.sys - won't. However, we alway sent request to specific database, otherwise queries may not work at all (currently I have errors when I sent requests without defined database).

I think it's better to remove ternaries - they have no sense.

Also I found more complex issue, we don't determine db correctly for nested entities, so for nested databases we have wrong db anyway. So actually, there is no difference, whether we pass database or not, .sys path could be relative when db is passed. Probably I revert the changes where I add database to .sys path

[artemmufazalov]

Removed unneeded ternaries, database should never be empty or request won't be successful anyway. BTW, these ternaries were not needed even in case of empty string database, YQL processed it correctly, they were needed for the case of undefined database.

Used relative paths everywhere: both in query text and api request we use the same param, so it never differs. If we use relative path, the query will use path from the database, that we passed.

[astandrik]

As I see here path and database can exist simulteneously
Could you please explain the real difference (as somewhere we change path to database and somewhere not)

[artemmufazalov]

Example: we look at shards of a table db1/table1
Database is db1, it should be used as a request param, system tables in db1/.sys/, but the result should be filtered, so only shards of current path db1/table1 will be shown

[artemmufazalov]

path - path to schema object
database - database, where this schema object exists

[sareyu]

LGTM