Failed task scheduling based on driver capability

[sushanthakumar]

What this PR does / why we need it:
This PR addresses the failed task scheduling mechanism considering the driver capability about performance metric retention window.

Which issue this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close that issue when PR gets merged): fixes #
This PR addresses the issue #757

Special notes for your reviewer:
Test scenarios verified here are:

Driver supports performance_metric_retention_window which is less than delfin max collection window
Driver supports performance_metric_retention_window which is more than delfin max collection window
Driver does not support performance_metric_retention_window -> use default interval
Driver throws exception for get_driver_capabilites

Release note:

[codecov]

Codecov Report

Merging #763 (4ff906c) into master (b8de7c6) will decrease coverage by 0.00%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master     #763      +/-   ##
==========================================
- Coverage   70.42%   70.41%   -0.01%     
==========================================
  Files         163      163              
  Lines       15836    15850      +14     
  Branches     1960     1961       +1     
==========================================
+ Hits        11152    11161       +9     
- Misses       4010     4014       +4     
- Partials      674      675       +1     

Impacted Files	Coverage Δ
delfin/api/schemas/storage_capabilities_schema.py	100.00% <ø> (ø)
delfin/common/config.py	95.00% <ø> (ø)
delfin/drivers/fake_storage/__init__.py	95.06% <ø> (ø)
...dulers/telemetry/performance_collection_handler.py	92.64% <64.28%> (-7.36%)	⬇️
delfin/common/constants.py	100.00% <100.00%> (ø)

[sushanthakumar]

@NajmudheenCT @joseph-v @ThisIsClark pls help to review this

[joseph-v]

LGTM

[NajmudheenCT]

LGTM

[ThisIsClark]

LGTM