Refactor transform scripts to drop temporary tables in a centralized way

[burrowse]

Within the current transform, temporary tables are created and dropped in various parts in the script making it difficult to test changes that may create temporary tables or assert data temporarily before loading in the "final" tables.

Examples (including but not limited to) are as follows:

h_immunization.sql

pedsnetcdm_to_pcornetcdm/v4.3_to_v6.0/sql_etl/scripts/etl_scripts/h_immunization.sql

Lines 51 to 53 in e79db62

	begin;
	drop table SITE_pcornet.imm_code;
	commit;

https://github.com/PEDSnet/pedsnetcdm_to_pcornetcdm/blob/e79db62fbb902476abbf43295de3a5256923948c/v4.3_to_v6.0/sql_etl/scripts/etl_scripts/h_immunization.sql#L69-71

n_obs_clin.sql

pedsnetcdm_to_pcornetcdm/v4.3_to_v6.0/sql_etl/scripts/etl_scripts/n_obs_clin.sql

Lines 174 to 177 in 31b7b27

	drop table if exists SITE_pcornet.meas_obsclin;
	drop table if exists SITE_pcornet.obs_vaping;
	drop table if exists SITE_pcornet.tobacco_obclin;
	drop table if exists SITE_pcornet.gestational_age;

Potential solution is to move these to the current drop table script : r_drop_tbl so all clean-up steps can be performed in one place.