[dbnode] Remove implicit cloning of time ranges to reduce Allocs

[notbdu]

What this PR does / why we need it:

Fixes #2170 . Removes implicit cloning of time ranges. Adds explicit cloning to the API.

Special notes for your reviewer:

Does this PR introduce a user-facing and/or backwards incompatible change?:

NONE

Does this PR require updating code package or user-facing documentation?:

NONE

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@fb46ae7). Click here to learn what that means.
The diff coverage is 85%.

@@           Coverage Diff            @@
##             master   #2178   +/-   ##
========================================
  Coverage          ?   72.2%           
========================================
  Files             ?    1019           
  Lines             ?   88456           
  Branches          ?       0           
========================================
  Hits              ?   63907           
  Misses            ?   20255           
  Partials          ?    4294

Flag	Coverage Δ
#aggregator	82% <ø> (?)
#cluster	85.3% <ø> (?)
#collector	82.8% <ø> (?)
#dbnode	79% <82.1%> (?)
#m3em	74.4% <ø> (?)
#m3ninx	74.7% <ø> (?)
#m3nsch	51.1% <ø> (?)
#metrics	17.6% <ø> (?)
#msg	74.8% <ø> (?)
#query	68.1% <ø> (?)
#x	83.1% <100%> (?)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update fb46ae7...2a8b608. Read the comment docs.

[robskillington]

Can we make this Get(shard uint32) (xtime.Ranges, bool)?

The amount of times we should have tested whether Get returns nil or not but forgot and then it panics (because someone didn't check whether it was nil before calling on it) is just too high to avoid making this change IMO.

[robskillington]

LGTM