Improve API of clientarch

[fernandohds564]

To be merged after #1230.

This PR changes the interface of the classes PVData, PVDataSet and ClientArchiver from siriuspy.clientarch package.

The main modifications are:

• renaming of all methods of ClientArchiver to snake_case, the python standard;
• addition of new funcionalities in ClientArchiver:
  • login now takes the password as an optional argument. If not provided, it will prompt the user for the password, in a discrete way. The old interface required the user to write its password in scripts and or ipython terminals, which would result in sensitive data being saved in files and logged in histories, etc...
  • get_data now supports all types of processing the archiver interface provides, not just the 'mean'. It takes two parameters, namely proc_type_param1 and proc_type_param2 to control the parameters of the chosen statistics. Some statistics may use one or none of them;
  • get_report was created and supports all types of reports provided by the archiver interface;
  • get_detailed_appliance_metrics was created and provides the results of the 'getApplianceMetricsForAppliance' call to archiver;
  • get_process_metrics_for_appliance was created and provides the results of the 'getProcessMetricsDataForAppliance';
  • the property timeout was renamed to query_timeout to better reflect its meaning: the timeout of each query transaction, instead of the total timeout of all queries.
  • the property parallel_query_bin_interval was renamed to query_split_interval. Now it will always be taken into consideration when creating queries urls, for splitting large time intervals into smaller ones;
  • the option parallel of get_data was removed. we can achieve similar results by setting query_split_interval to 0;
  • the property query_max_concurrency was created to limit the maximum number of concurrent queries that will be requested from the server. It is an additional control for optimizing the retrieval time.
• changes/improvements of PVData API:
  • implementation of __str__ method to show a nice visualization of the class properties
  • removal of timestamp_start and timestamp_stop. Now only time_start and time_stop exists. Their getter return a clientarch.Time but their setters accept anything that can be converted to Time object, including timestamps;
  • the properties processing_type, processing_type_param1, processing_type_param2 and query_split_interval were added to control how self.connector.get_data will be called inside the self.update method. This means the self.update method does not take any arguments beside a query_timeout, that overwrites self.connector.query_timeout if passed;
  • the properties query_max_concurrency and query_timeout were added to ease setting of the self.connector respective properties;
• Changes/improvements of PVDataSet API:
  • implementation of __str__ method to show a nice visualization of the class properties
  • removal of timestamp_start and timestamp_stop. Now only time_start and time_stop exists. Their getter return a clientarch.Time if all internal PVData objects have equal start_time|stop_time or a list of them, one for each PVData object. Their setters accept anything that can be converted to Time object, including timestamps. It also accepts a list of Time objects, referring to the desired start_time|stop_time of each PVData.
  • the properties processing_type, processing_type_param1, processing_type_param2 and query_split_interval were added to control how self.connector.get_data will be called inside the self.update method. Their getters return a single value or a list of values depending if the associated PVData's are identical or not, and the setter also accepts a list of values, one for each PVData. This means the self.update method does not take any arguments beside a query_timeout, that overwrites self.connector.query_timeout if passed;
  • the properties query_max_concurrency and query_timeout were added to ease setting of the self.connector respective properties;

This PR was testes in the control room computers, using the script gensumm_machreport.py and other manual tests.