Update run_id when adding ds in mem to db

[Dominik-Vogel]

This should fix the issue of run_ids not getting incremented when importing netcdf datasets or datasets in memory in general.
Addressing the issue we have talked about @jenshnielsen .

I have replaced the table name return value, as I could not find it used anywhere and it is not a part of a public API.

[jenshnielsen]

@Dominik-Vogel Thanks. Looks like it is used in database_extract_runs.py see typechecking error.

I think we should see if we can write a test for this too

[codecov]

Codecov Report

Merging #5209 (270b302) into master (c5ec9c7) will decrease coverage by 0.02%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #5209      +/-   ##
==========================================
- Coverage   67.36%   67.34%   -0.02%     
==========================================
  Files         359      359              
  Lines       29896    29901       +5     
==========================================
- Hits        20138    20136       -2     
- Misses       9758     9765       +7     

[jenshnielsen]

@Dominik-Vogel I have added tests and implemented the same fix for counter and exp_id. Could you have another look?